MatlabOnly.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Setup\n",
    "Setup MIRT and \"reproduce-l-s-dynamic-mri\" matlab projects\n",
    "\n",
    " - https://github.com/JeffFessler/mirt\n",
    " - https://github.com/JeffFessler/reproduce-l-s-dynamic-mri\n",
    "   - And it is based on Ricardo Otazo's code: http://cai2r.net/resources/software/ls-reconstruction-matlab-code"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Note: in order to make this notebook work, I installed Matlab's python engine (https://www.mathworks.com/help/matlab/matlab_external/install-matlab-engine-api-for-python-in-nondefault-locations.html), and also a Matlab kernel for Jupyter (https://github.com/Calysto/matlab_kernel)*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The variable \"irtdir\" is not set, so trying default, assuming\n",
      "that you launched matlab from the irt install directory.\n",
      "You may need to edit setup.m or adjust your path otherwise.\n",
      "Assuming you installed irt in directory \"/home/hakkelt/mri/mirt\".\n",
      "Warning: Name is nonexistent or not a directory: /home/hakkelt/mri/mirt/mex/v7\n",
      "> In path (line 109)\n",
      "  In addpath (line 86)\n",
      "  In setup (line 70)\n",
      "  In run (line 91)\n",
      "Path setup for irt appears to have succeeded.\n"
     ]
    }
   ],
   "source": [
    "addpath('../mirt')\n",
    "run('../mirt/setup')\n",
    "addpath('../reproduce-l-s-dynamic-mri/data')\n",
    "addpath('../reproduce-l-s-dynamic-mri/toolbox_pincat/')\n",
    "addpath('../reproduce-l-s-dynamic-mri/operators/')\n",
    "addpath('../reproduce-l-s-dynamic-mri/algorithms/')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. PINCAT phantom example\n",
    "https://github.com/JeffFessler/reproduce-l-s-dynamic-mri/blob/master/examples/example_PINCAT_phantom.m"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.1 Preparation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Read data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "% L+S reconstruction of undersampled multicoil PINCAT phantom\n",
    "% Claire Lin, 06/05/2018\n",
    "clear all;close all;\n",
    "load('data/aperiodic_pincat.mat')\n",
    "load('data/Xinf.mat')\n",
    "tmp = max(new(:));\n",
    "Xtrue = div0(new,tmp);\n",
    "[nx,ny,nt] = size(Xtrue);\n",
    "im(Xtrue)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Simulate coil maps**\n",
    "\n",
    "Code for `ir_mri_sensemap_sim`: https://github.com/JeffFessler/reproduce-l-s-dynamic-mri/blob/master/toolbox_pincat/ir_mri_sensemap_sim.m"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "warn im: magnitude of complex image\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%% simulate coil maps\n",
    "% 4 rings of 8 coils\n",
    "nc = 8;\n",
    "nring = 4;\n",
    "b1 = ir_mri_sensemap_sim('chat', 0, 'nx', nx, 'ny', 1, 'nz', ny,'dx', 1.5, 'dz', 1.5, ...\n",
    "\t'rcoil', 120, 'nring', nring, 'ncoil', nc*nring, 'coil_distance', 1.2);\n",
    "tmp = sqrt(sum(abs((b1)).^2,3));\n",
    "b1c = div0(b1,tmp);\n",
    "im(b1c)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Coil compression to 8 coils**\n",
    "\n",
    "MRI coil compression via PCA<br>\n",
    "Given multiple MRI surface coil images (b1c), use SVD/PCA to find a smaller number of virtual coil images."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "warn im: magnitude of complex image\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%% coil compression to 8 coils\n",
    "[smap0, ~] = ir_mri_coil_compress(b1c, 'ncoil', nc);\n",
    "tmp = sqrt(sum(abs((smap0)).^2,3));\n",
    "smap = div0(smap0,tmp);\n",
    "im(smap)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Sampling mask**\n",
    "\n",
    "Code for `strucrand` can be found here: https://github.com/JeffFessler/reproduce-l-s-dynamic-mri/blob/master/toolbox_pincat/strucrand.m"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%% sampling mask\n",
    "line = 24;\n",
    "samp = strucrand(nx,ny,nt,line);\n",
    "im(samp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Acquisition operator and measured data**\n",
    "\n",
    "Code for `getE`: https://github.com/JeffFessler/reproduce-l-s-dynamic-mri/blob/master/operators/getE.m"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "%% data\n",
    "E=getE(smap,nt,'samp',samp);\n",
    "d = E*Xtrue;"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Add noise and prepare for regularization scaling**\n",
    "\n",
    "Add complex noise such that the $SNR = 46dB$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ": data rmse = 0.000425417, snr = 46\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "% add noise\n",
    "rng(0)\n",
    "dn = randn(size(d)) + 1i * randn(size(d));\n",
    "param.snr_db = 46;\n",
    "param.scale_noise = norm(d(:)) / norm(dn(:)) / 10.^(param.snr_db / 20);\n",
    "param.d = d + param.scale_noise * dn;\n",
    "printm('data rmse = %g, snr = %g', rms(param.d(:)-d(:)), ...\n",
    "    20*log10(norm(d(:)) / norm(param.d(:)-d(:))))\n",
    "opt.d = param.d;\n",
    "\n",
    "% prepare for regularization scaling\n",
    "L = E'*param.d; \n",
    "res = E*L-param.d;\n",
    "[~,St,~]=svd(reshape(L,[nx*ny,nt])-reshape(E'*res,[nx*ny,nt]),0);\n",
    "semilogy(diag(St),'ok')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.2 Test algorithms"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Prepare for augmented Lagrangian algorithms**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "%% prepare for AL: opt\n",
    "opt.smap = smap;\n",
    "opt.T=getT(nx,ny,nt);\n",
    "opt.nite = 10;\n",
    "opt.samp = samp;\n",
    "[opt.F,opt.C] = getFS(opt.smap,nt,'samp',opt.samp);\n",
    "opt.E=getE(smap,nt,'samp',opt.samp);\n",
    "opt.scaleL = St(1);\n",
    "opt.scaleS = 1/1.887; %1 / b1 constant squared in middle of image\n",
    "opt.muL=0.01;\n",
    "opt.muS=0.05*opt.scaleS;\n",
    "opt.Xinf = Xinf.pincat;"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run AL-CG**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 25.3092403, cost: 951.1688233, rank of L: 50\n",
      " ********** AL-CG: L+S reconstruction **********\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 3.3e-05.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00035.\n",
      " ite: 1 , rank of L: 49, xdiff: 13.7454393, cost: 863.6173103\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 1.8e-05.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00032.\n",
      " ite: 2 , rank of L: 25, xdiff: 10.4156443, cost: 816.3825683\n",
      "cgs converged at iteration 3 to a solution with relative residual 9.7e-06.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0002.\n",
      " ite: 3 , rank of L: 15, xdiff: 8.3161273, cost: 792.2092293\n",
      "cgs converged at iteration 3 to a solution with relative residual 5.6e-06.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0002.\n",
      " ite: 4 , rank of L: 11, xdiff: 7.1471373, cost: 766.1499023\n",
      "cgs converged at iteration 3 to a solution with relative residual 4e-06.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00014.\n",
      " ite: 5 , rank of L: 12, xdiff: 6.4871593, cost: 754.0796513\n",
      "cgs converged at iteration 3 to a solution with relative residual 3.6e-06.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 8.6e-05.\n",
      " ite: 6 , rank of L: 13, xdiff: 6.0405853, cost: 748.2368773\n",
      "cgs converged at iteration 3 to a solution with relative residual 3.1e-06.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 5.9e-05.\n",
      " ite: 7 , rank of L: 12, xdiff: 5.7215223, cost: 743.8823853\n",
      "cgs converged at iteration 3 to a solution with relative residual 2.5e-06.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 4.6e-05.\n",
      " ite: 8 , rank of L: 10, xdiff: 5.5022403, cost: 741.0316773\n",
      "cgs converged at iteration 3 to a solution with relative residual 1.9e-06.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 3.9e-05.\n",
      " ite: 9 , rank of L: 10, xdiff: 5.3463433, cost: 737.6754763\n",
      "cgs converged at iteration 3 to a solution with relative residual 1.5e-06.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 3.4e-05.\n",
      " ite: 10 , rank of L: 9, xdiff: 5.2255333, cost: 733.3815923\n",
      "\n",
      "elapsed time: 16.8064\n"
     ]
    }
   ],
   "source": [
    "%% AL-CG\n",
    "d1 = 1/2; d2 = 1/2; %for AL-CG\n",
    "c = clock();\n",
    "[L_cg,S_cg,x_cg,cost_cg,time_cg,rankL_cg] = AL_CG(opt,'d1',d1,'d2',d2);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run AL-2**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 25.3092403, cost: 951.1688233, rank of L: 50\n",
      " ********** AL-2: L+S reconstruction **********\n",
      " ite: 1 , rank of L: 3, xdiff: 41.9408913, cost: 1417.4295653\n",
      " ite: 2 , rank of L: 3, xdiff: 26.8228993, cost: 961.5783083\n",
      " ite: 3 , rank of L: 4, xdiff: 18.7992523, cost: 816.4223633\n",
      " ite: 4 , rank of L: 6, xdiff: 12.6015103, cost: 747.2053223\n",
      " ite: 5 , rank of L: 7, xdiff: 9.3126613, cost: 724.4708253\n",
      " ite: 6 , rank of L: 7, xdiff: 7.8415493, cost: 717.4979863\n",
      " ite: 7 , rank of L: 8, xdiff: 6.7806903, cost: 712.8419803\n",
      " ite: 8 , rank of L: 8, xdiff: 6.0018883, cost: 709.7363283\n",
      " ite: 9 , rank of L: 8, xdiff: 5.5401633, cost: 708.1655273\n",
      " ite: 10 , rank of L: 8, xdiff: 5.1972273, cost: 707.1064453\n",
      "\n",
      "elapsed time: 17.8129\n"
     ]
    }
   ],
   "source": [
    "%% AL-2\n",
    "d1 = 1/3; d2 = 1/10; %for AL-2\n",
    "c = clock();\n",
    "[L_al,S_al,xdiff_al,cost_al,time_al,rankL_al] = AL_2(opt,'d1',d1,'d2',d2);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Prepare for PGM algorithms**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "%% prepare for PGM: param\n",
    "param.E = getE(smap,nt,'samp',samp);\n",
    "param.T = getT(nx,ny,nt);\n",
    "param.nite = 10;\n",
    "param.scaleL = St(1);\n",
    "param.scaleS = 1/1.887; %1 / b1 constant squared in middle of image\n",
    "param.lambda_L=0.01;\n",
    "param.lambda_S=0.05*param.scaleS;\n",
    "param.Xinf = reshape(Xinf.pincat,nx*ny,nt);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run ISTA**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 25.3092403, cost: 951.1688233, rank of L: 50\n",
      " ********** PGM: L+S reconstruction **********\n",
      " ite: 1, rank of L: 20, xdiff: 15.8493393, cost: 786.3693853\n",
      " ite: 2, rank of L: 15, xdiff: 11.1977013, cost: 752.6289673\n",
      " ite: 3, rank of L: 13, xdiff: 9.0156423, cost: 738.7617193\n",
      " ite: 4, rank of L: 12, xdiff: 7.5550733, cost: 732.6481933\n",
      " ite: 5, rank of L: 12, xdiff: 6.9909103, cost: 729.1419683\n",
      " ite: 6, rank of L: 12, xdiff: 6.4011543, cost: 726.5877693\n",
      " ite: 7, rank of L: 11, xdiff: 6.2127023, cost: 724.5297243\n",
      " ite: 8, rank of L: 11, xdiff: 5.8614373, cost: 722.7829593\n",
      " ite: 9, rank of L: 11, xdiff: 5.7729363, cost: 721.2694093\n",
      " ite: 10, rank of L: 10, xdiff: 5.5149013, cost: 719.9468993\n",
      "\n",
      "elapsed time: 1.7272\n"
     ]
    }
   ],
   "source": [
    "%% ISTA\n",
    "c = clock();\n",
    "[L_ista,S_ista,xdiff_ista,cost_ista,time_ista,rankL_ista] = PGM(param);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run FISTA**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 25.3092403, cost: 951.1688233, rank of L: 50\n",
      " ********** PGM: L+S reconstruction **********\n",
      " ite: 1, rank of L: 34, xdiff: 19.3698183, cost: 831.9344483\n",
      " ite: 2, rank of L: 23, xdiff: 15.8861963, cost: 787.8776253\n",
      " ite: 3, rank of L: 18, xdiff: 12.7468273, cost: 761.8648073\n",
      " ite: 4, rank of L: 15, xdiff: 10.2047963, cost: 745.7527473\n",
      " ite: 5, rank of L: 13, xdiff: 8.3865943, cost: 736.0908813\n",
      " ite: 6, rank of L: 12, xdiff: 7.2166353, cost: 730.9735113\n",
      " ite: 7, rank of L: 12, xdiff: 6.5144553, cost: 727.4960943\n",
      " ite: 8, rank of L: 11, xdiff: 6.0735573, cost: 724.6638793\n",
      " ite: 9, rank of L: 11, xdiff: 5.7702753, cost: 722.1716313\n",
      " ite: 10, rank of L: 10, xdiff: 5.5237363, cost: 719.9782713\n",
      "\n",
      "elapsed time: 2.4069\n"
     ]
    }
   ],
   "source": [
    "%% FISTA\n",
    "c = clock();\n",
    "[L_fista,S_fista,xdiff_fista,cost_fista,time_fista,rankL_fista] = PGM(param,'fistaL',1,'fistaS',1);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run POGM**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 25.3092403, cost: 951.1688233, rank of L: 50\n",
      " ********** PGM: L+S reconstruction **********\n",
      " ite: 1, rank of L: 25, xdiff: 16.8706053, cost: 796.9192503\n",
      " ite: 2, rank of L: 15, xdiff: 11.4006803, cost: 753.0890503\n",
      " ite: 3, rank of L: 12, xdiff: 8.1827443, cost: 735.4837653\n",
      " ite: 4, rank of L: 12, xdiff: 6.6846283, cost: 728.9039923\n",
      " ite: 5, rank of L: 11, xdiff: 6.0296173, cost: 724.5433353\n",
      " ite: 6, rank of L: 10, xdiff: 5.6109743, cost: 720.9524543\n",
      " ite: 7, rank of L: 9, xdiff: 5.3057743, cost: 717.9908453\n",
      " ite: 8, rank of L: 9, xdiff: 5.0435533, cost: 715.4371953\n",
      " ite: 9, rank of L: 9, xdiff: 4.8189883, cost: 713.3032843\n",
      " ite: 10, rank of L: 9, xdiff: 4.6525973, cost: 711.9748543\n",
      "\n",
      "elapsed time: 1.7870\n"
     ]
    }
   ],
   "source": [
    "%% POGM\n",
    "c = clock();\n",
    "[L_pogm,S_pogm,xdiff_pogm,cost_pogm,time_pogm,rankL_pogm] = PGM(param,'pogmS',1,'pogmL',1);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3. Multicoil cardiac cine MRI\n",
    "https://github.com/JeffFessler/reproduce-l-s-dynamic-mri/blob/master/examples/example_cardiac_cine.m"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.1 Preparation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Read data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "% L+S reconstruction of undersampled multicoil cardiac cine MRI\n",
    "% Claire Lin, 05/20/2018\n",
    "clear all;close all;\n",
    "load('data/cardiac_cine_R6.mat')\n",
    "load ('data/Xinf.mat')\n",
    "%% normalize smap\n",
    "[nx,ny,nt,nc]=size(kdata);\n",
    "tmp = sqrt(sum(abs((b1)).^2,3));\n",
    "b1c = div0(b1,tmp);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.2 Test algorithms"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Prepare for augmented Lagrangian methods**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "%% prepare for AL: opt\n",
    "opt.d = kdata;\n",
    "opt.smap = b1c;\n",
    "opt.T=getT(nx,ny,nt);\n",
    "opt.nite = 10;\n",
    "opt.samp = opt.d(:,:,:,1)~=0;\n",
    "[opt.F,opt.C] = getFS(opt.smap,nt,'samp',opt.samp);\n",
    "opt.E=getE(b1c,nt,'samp',opt.samp);\n",
    "% scalers to match Otazo's results\n",
    "opt.scaleL = 175/1.2656; % Otazo's stopping St(1) / b1 constant squared\n",
    "opt.scaleS = 1/1.2656; % 1 / b1 constant squared\n",
    "opt.muL=0.01;\n",
    "opt.muS=0.0025*opt.scaleS;\n",
    "opt.Xinf = Xinf.cine;"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run AL-CG**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 111.1760183, cost: 2056.6478743, rank of L: 24\n",
      " ********** AL-CG: L+S reconstruction **********\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0034.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.036.\n",
      " ite: 1 , rank of L: 24, xdiff: 51.1746563, cost: 682.8310873\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0026.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.016.\n",
      " ite: 2 , rank of L: 24, xdiff: 28.9228803, cost: 574.8587193\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0008.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.012.\n",
      " ite: 3 , rank of L: 24, xdiff: 22.8716033, cost: 545.4449573\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00054.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.009.\n",
      " ite: 4 , rank of L: 24, xdiff: 19.3625373, cost: 525.9914013\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00044.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0069.\n",
      " ite: 5 , rank of L: 24, xdiff: 17.0936323, cost: 507.5359933\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00035.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.005.\n",
      " ite: 6 , rank of L: 24, xdiff: 15.8326003, cost: 496.5016503\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00031.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0038.\n",
      " ite: 7 , rank of L: 24, xdiff: 15.1447563, cost: 491.9777453\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00029.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0031.\n",
      " ite: 8 , rank of L: 24, xdiff: 14.7147543, cost: 490.2126043\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00027.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0027.\n",
      " ite: 9 , rank of L: 24, xdiff: 14.3750633, cost: 488.5522583\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00024.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0023.\n",
      " ite: 10 , rank of L: 24, xdiff: 14.0499873, cost: 484.0706813\n",
      "\n",
      "elapsed time: 49.0223\n"
     ]
    }
   ],
   "source": [
    "%% AL-CG\n",
    "d1 = 1/10; d2 = 1/20; %for AL-CG\n",
    "c = clock();\n",
    "[L_cg,S_alcg,x_alcg,cost_alcg,time_alcg,rankL_alcg] = AL_CG(opt,'d1',d1,'d2',d2);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run AL-2**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 111.1760183, cost: 2056.6478743, rank of L: 24\n",
      " ********** AL-2: L+S reconstruction **********\n",
      " ite: 1 , rank of L: 1, xdiff: 95.6885453, cost: 2365.8748573\n",
      " ite: 2 , rank of L: 1, xdiff: 65.7433623, cost: 798.3235803\n",
      " ite: 3 , rank of L: 0, xdiff: 53.3320393, cost: 413.0994443\n",
      " ite: 4 , rank of L: 0, xdiff: 46.7413293, cost: 323.3674703\n",
      " ite: 5 , rank of L: 0, xdiff: 41.2476313, cost: 289.0024733\n",
      " ite: 6 , rank of L: 0, xdiff: 36.6471563, cost: 269.1716313\n",
      " ite: 7 , rank of L: 0, xdiff: 33.0463753, cost: 258.1183483\n",
      " ite: 8 , rank of L: 0, xdiff: 30.2602373, cost: 251.8830083\n",
      " ite: 9 , rank of L: 0, xdiff: 28.1066283, cost: 248.0244133\n",
      " ite: 10 , rank of L: 0, xdiff: 26.4248773, cost: 245.1701213\n",
      "\n",
      "elapsed time: 55.2986\n"
     ]
    }
   ],
   "source": [
    "%% AL-2\n",
    "d1 = 1/10; d2 = 1/100;  %for AL-2\n",
    "c = clock();\n",
    "[L_al2,S_al2,xdiff_al2,cost_al2,time_al2,rankL_al2] = AL_2(opt,'d1',d1,'d2',d2);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Prepare for PGM algorithms**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "%% prepare for PGM: param\n",
    "param.E=getE(b1c,nt,'samp',kdata(:,:,:,1)~=0);\n",
    "param.d=kdata;\n",
    "param.T = getT(nx,ny,nt);\n",
    "param.nite=10;\n",
    "param.scaleL = 175/1.2656;\n",
    "param.scaleS = 1/1.2656;\n",
    "param.lambda_L=0.01;\n",
    "param.lambda_S=0.0025*param.scaleS;\n",
    "param.Xinf = reshape(Xinf.cine,nx*ny,nt);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run ISTA**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 111.1760183, cost: 2056.6478743, rank of L: 24\n",
      " ********** PGM: L+S reconstruction **********\n",
      " ite: 1, rank of L: 24, xdiff: 77.2504503, cost: 931.0880953\n",
      " ite: 2, rank of L: 24, xdiff: 67.1571043, cost: 782.0542613\n",
      " ite: 3, rank of L: 24, xdiff: 60.1077463, cost: 709.3672743\n",
      " ite: 4, rank of L: 24, xdiff: 54.7802243, cost: 658.1694593\n",
      " ite: 5, rank of L: 23, xdiff: 50.2011303, cost: 617.5562673\n",
      " ite: 6, rank of L: 21, xdiff: 46.3694113, cost: 584.0904523\n",
      " ite: 7, rank of L: 19, xdiff: 42.9700513, cost: 556.4205143\n",
      " ite: 8, rank of L: 16, xdiff: 40.0760083, cost: 533.0474603\n",
      " ite: 9, rank of L: 15, xdiff: 37.5284423, cost: 513.8099193\n",
      " ite: 10, rank of L: 13, xdiff: 35.3816603, cost: 498.0421783\n",
      "\n",
      "elapsed time: 5.5376\n"
     ]
    }
   ],
   "source": [
    "%% ISTA\n",
    "c = clock();\n",
    "[L_ista,S_ista,xdiff_ista,cost_ista,time_ista,rankL_ista] = PGM(param);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run FISTA**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 111.1760183, cost: 2056.6478743, rank of L: 24\n",
      " ********** PGM: L+S reconstruction **********\n",
      " ite: 1, rank of L: 24, xdiff: 89.7275773, cost: 1203.0921193\n",
      " ite: 2, rank of L: 24, xdiff: 79.3396383, cost: 952.7407733\n",
      " ite: 3, rank of L: 24, xdiff: 71.1560133, cost: 821.6829323\n",
      " ite: 4, rank of L: 24, xdiff: 64.3450703, cost: 743.5552883\n",
      " ite: 5, rank of L: 24, xdiff: 58.3090363, cost: 686.0264113\n",
      " ite: 6, rank of L: 24, xdiff: 52.7729343, cost: 637.1883603\n",
      " ite: 7, rank of L: 23, xdiff: 47.6568833, cost: 593.8171413\n",
      " ite: 8, rank of L: 20, xdiff: 42.9705733, cost: 555.9679033\n",
      " ite: 9, rank of L: 16, xdiff: 38.7736433, cost: 523.7240033\n",
      " ite: 10, rank of L: 13, xdiff: 35.1568373, cost: 497.8138393\n",
      "\n",
      "elapsed time: 7.5572\n"
     ]
    }
   ],
   "source": [
    "%% FISTA\n",
    "c = clock();\n",
    "[L_fista,S_fista,xdiff_fista,cost_fista,time_fista,rankL_fista] = PGM(param,'fistaL',1,'fistaS',1);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run POGM**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 111.1760183, cost: 2056.6478743, rank of L: 24\n",
      " ********** PGM: L+S reconstruction **********\n",
      " ite: 1, rank of L: 24, xdiff: 80.6012043, cost: 968.3333903\n",
      " ite: 2, rank of L: 24, xdiff: 67.3216093, cost: 779.9223193\n",
      " ite: 3, rank of L: 24, xdiff: 57.6739773, cost: 686.7232793\n",
      " ite: 4, rank of L: 23, xdiff: 49.5012823, cost: 613.9794813\n",
      " ite: 5, rank of L: 19, xdiff: 42.4597663, cost: 555.2589473\n",
      " ite: 6, rank of L: 14, xdiff: 36.5725523, cost: 509.8669403\n",
      " ite: 7, rank of L: 8, xdiff: 32.0292363, cost: 479.2692883\n",
      " ite: 8, rank of L: 5, xdiff: 28.7907943, cost: 461.0693393\n",
      " ite: 9, rank of L: 3, xdiff: 26.4333953, cost: 449.8849533\n",
      " ite: 10, rank of L: 3, xdiff: 24.9634973, cost: 442.2500543\n",
      "\n",
      "elapsed time: 5.7645\n"
     ]
    }
   ],
   "source": [
    "%% POGM\n",
    "c = clock();\n",
    "[L_pogm,S_pogm,xdiff_pogm,cost_pogm,time_pogm,rankL_pogm] = PGM(param,'pogmS',1,'pogmL',1);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4. Multicoil cardiac perfusion MRI\n",
    "https://github.com/JeffFessler/reproduce-l-s-dynamic-mri/blob/master/examples/example_cardiac_perf.m"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.1 Preparation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Read data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "% L+S reconstruction of undersampled multicoil cardiac perfusion MRI\n",
    "% Claire Lin, 05/20/2018\n",
    "clear all;close all;\n",
    "load('data/cardiac_perf_R8.mat')\n",
    "load('data/Xinf.mat')\n",
    "%% normalize smap\n",
    "[nx,ny,nt,nc]=size(kdata);\n",
    "tmp = sqrt(sum(abs((b1)).^2,3));\n",
    "b1c = div0(b1,tmp);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.2 Test algorithms"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Prepare for Augmented Lagrangian algorithms**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "%% prepare for AL: opt\n",
    "opt.d = kdata;\n",
    "opt.smap = b1c;\n",
    "opt.T=getT(nx,ny,nt);\n",
    "opt.nite = 10;\n",
    "opt.samp = opt.d(:,:,:,1)~=0;\n",
    "[opt.F,opt.C] = getFS(opt.smap,nt,'samp',opt.samp);\n",
    "opt.E=getE(b1c,nt,'samp',opt.samp);\n",
    "% scalers to match Otazo's results\n",
    "opt.scaleL = 130/1.2775; % Otazo's stopping St(1) / b1 constant squared\n",
    "opt.scaleS = 1/1.2775; % 1 / b1 constant squared\n",
    "opt.muL=0.01;\n",
    "opt.muS=0.01*opt.scaleS;\n",
    "opt.Xinf = Xinf.perf;"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run AL-CG**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 118.9517593, cost: 1605.9090993, rank of L: 40\n",
      " ********** AL-CG: L+S reconstruction **********\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0018.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0068.\n",
      " ite: 1 , rank of L: 40, xdiff: 81.4396513, cost: 677.7361773\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00054.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0045.\n",
      " ite: 2 , rank of L: 40, xdiff: 63.3424423, cost: 580.3501733\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00047.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0028.\n",
      " ite: 3 , rank of L: 40, xdiff: 51.3357123, cost: 527.5888823\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0004.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0021.\n",
      " ite: 4 , rank of L: 40, xdiff: 42.4700203, cost: 497.5957143\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00032.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0016.\n",
      " ite: 5 , rank of L: 40, xdiff: 35.9206853, cost: 480.0069793\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00025.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0013.\n",
      " ite: 6 , rank of L: 40, xdiff: 31.2473773, cost: 468.7550633\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0002.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.0011.\n",
      " ite: 7 , rank of L: 40, xdiff: 28.0101413, cost: 460.2363773\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00017.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00094.\n",
      " ite: 8 , rank of L: 40, xdiff: 25.7848783, cost: 453.2828203\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00014.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00079.\n",
      " ite: 9 , rank of L: 40, xdiff: 24.2571303, cost: 448.0839923\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00011.\n",
      "cgs stopped at iteration 3 without converging to the desired tolerance 1e-05\n",
      "because the maximum number of iterations was reached.\n",
      "The iterate returned (number 3) has relative residual 0.00067.\n",
      " ite: 10 , rank of L: 40, xdiff: 23.2105733, cost: 444.4904893\n",
      "\n",
      "elapsed time: 19.9059\n"
     ]
    }
   ],
   "source": [
    "%% AL-CG\n",
    "d1 = 1/5; d2 = 1/5; %for AL-CG\n",
    "c = clock();\n",
    "[L_cg,S_cg,x_cg,cost_cg,time_cg,rankL_cg] = AL_CG(opt,'d1',d1,'d2',d2);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run AL-2**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 118.9517593, cost: 1605.9090993, rank of L: 40\n",
      " ********** AL-2: L+S reconstruction **********\n",
      " ite: 1 , rank of L: 1, xdiff: 118.0958713, cost: 2919.1098513\n",
      " ite: 2 , rank of L: 1, xdiff: 89.0041583, cost: 1310.4811123\n",
      " ite: 3 , rank of L: 3, xdiff: 77.6117863, cost: 820.5040173\n",
      " ite: 4 , rank of L: 3, xdiff: 70.6298073, cost: 642.2338453\n",
      " ite: 5 , rank of L: 4, xdiff: 64.0606163, cost: 552.3495333\n",
      " ite: 6 , rank of L: 5, xdiff: 58.0906793, cost: 500.7592743\n",
      " ite: 7 , rank of L: 5, xdiff: 52.7845843, cost: 469.7612703\n",
      " ite: 8 , rank of L: 4, xdiff: 48.1143843, cost: 450.6305853\n",
      " ite: 9 , rank of L: 4, xdiff: 44.0771873, cost: 438.6039523\n",
      " ite: 10 , rank of L: 4, xdiff: 40.6270413, cost: 430.6803203\n",
      "\n",
      "elapsed time: 27.5002\n"
     ]
    }
   ],
   "source": [
    "%% AL-2\n",
    "d1 = 1/5; d2 = 1/50; %for AL-2\n",
    "c = clock();\n",
    "[L_al,S_al,xdiff_al,cost_al,time_al,rankL_al] = AL_2(opt,'d1',d1,'d2',d2);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Prepare for PGM algorithms**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "%% prepare for PGM: param\n",
    "param.E=getE(b1c,nt,'samp',kdata(:,:,:,1)~=0);\n",
    "param.d=kdata;\n",
    "param.T = getT(nx,ny,nt);\n",
    "param.nite=10;\n",
    "param.scaleL = 130/1.2775;\n",
    "param.scaleS = 1/1.2775;\n",
    "param.lambda_L=0.01;\n",
    "param.lambda_S=0.01*param.scaleS;\n",
    "param.Xinf = reshape(Xinf.perf,nx*ny,nt);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run ISTA**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 118.9517593, cost: 1605.9090993, rank of L: 40\n",
      " ********** PGM: L+S reconstruction **********\n",
      " ite: 1, rank of L: 40, xdiff: 94.7225423, cost: 819.4575223\n",
      " ite: 2, rank of L: 40, xdiff: 86.7924423, cost: 707.9456013\n",
      " ite: 3, rank of L: 40, xdiff: 80.7610023, cost: 651.7363823\n",
      " ite: 4, rank of L: 40, xdiff: 75.9018943, cost: 612.8571323\n",
      " ite: 5, rank of L: 38, xdiff: 71.5280463, cost: 583.2456303\n",
      " ite: 6, rank of L: 36, xdiff: 67.6884923, cost: 559.7511453\n",
      " ite: 7, rank of L: 33, xdiff: 64.1339653, cost: 540.6652673\n",
      " ite: 8, rank of L: 29, xdiff: 60.9227943, cost: 524.8275273\n",
      " ite: 9, rank of L: 28, xdiff: 57.9282653, cost: 511.4539013\n",
      " ite: 10, rank of L: 27, xdiff: 55.1851813, cost: 499.9599543\n",
      "\n",
      "elapsed time: 2.2775\n"
     ]
    }
   ],
   "source": [
    "%% ISTA\n",
    "c = clock();\n",
    "[L_ista,S_ista,xdiff_ista,cost_ista,time_ista,rankL_ista] = PGM(param);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run FISTA**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 118.9517593, cost: 1605.9090993, rank of L: 40\n",
      " ********** PGM: L+S reconstruction **********\n",
      " ite: 1, rank of L: 40, xdiff: 103.8934403, cost: 1016.8542883\n",
      " ite: 2, rank of L: 40, xdiff: 96.2738493, cost: 836.7154403\n",
      " ite: 3, rank of L: 40, xdiff: 89.9116443, cost: 738.6073813\n",
      " ite: 4, rank of L: 40, xdiff: 84.2732853, cost: 678.2913103\n",
      " ite: 5, rank of L: 40, xdiff: 78.9840853, cost: 633.9038893\n",
      " ite: 6, rank of L: 40, xdiff: 73.8725053, cost: 597.3557363\n",
      " ite: 7, rank of L: 37, xdiff: 68.8839953, cost: 566.2765563\n",
      " ite: 8, rank of L: 33, xdiff: 64.0132223, cost: 539.8553613\n",
      " ite: 9, rank of L: 29, xdiff: 59.2761233, cost: 517.4710143\n",
      " ite: 10, rank of L: 27, xdiff: 54.7063873, cost: 498.4585513\n",
      "\n",
      "elapsed time: 3.1283\n"
     ]
    }
   ],
   "source": [
    "%% FISTA\n",
    "c = clock();\n",
    "[L_fista,S_fista,xdiff_fista,cost_fista,time_fista,rankL_fista] = PGM(param,'fistaL',1,'fistaS',1);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run POGM**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 118.9517593, cost: 1605.9090993, rank of L: 40\n",
      " ********** PGM: L+S reconstruction **********\n",
      " ite: 1, rank of L: 40, xdiff: 97.3583763, cost: 849.4886573\n",
      " ite: 2, rank of L: 40, xdiff: 86.9139863, cost: 707.7278993\n",
      " ite: 3, rank of L: 40, xdiff: 78.6276933, cost: 635.4357403\n",
      " ite: 4, rank of L: 37, xdiff: 71.0008163, cost: 580.9478443\n",
      " ite: 5, rank of L: 32, xdiff: 63.7240603, cost: 539.0874053\n",
      " ite: 6, rank of L: 28, xdiff: 56.7864073, cost: 506.9395343\n",
      " ite: 7, rank of L: 22, xdiff: 50.2903563, cost: 481.9811793\n",
      " ite: 8, rank of L: 19, xdiff: 44.3746223, cost: 462.8262993\n",
      " ite: 9, rank of L: 16, xdiff: 39.1518293, cost: 448.2832673\n",
      " ite: 10, rank of L: 13, xdiff: 35.6455923, cost: 439.4732693\n",
      "\n",
      "elapsed time: 2.3336\n"
     ]
    }
   ],
   "source": [
    "%% POGM\n",
    "c = clock();\n",
    "[L_pogm,S_pogm,xdiff_pogm,cost_pogm,time_pogm,rankL_pogm] = PGM(param,'pogmS',1,'pogmL',1);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 5. Multicoil abdominal dce MRI\n",
    "https://github.com/JeffFessler/reproduce-l-s-dynamic-mri/blob/master/examples/example_abdomen_dce_ga.m"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.1 Preparation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Read data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "% L+S reconstruction of undersampled multicoil abdominal dce MRI\n",
    "% Claire Lin, 05/07/2019\n",
    "%clear all;close all;\n",
    "load('data/abdomen_dce_ga.mat')\n",
    "load ('data/Xinf.mat')\n",
    "%% number of spokes to be used per frame (Fibonacci number)\n",
    "nspokes=21;\n",
    "[nx,ny,nc]=size(b1);\n",
    "[nr,ntviews,nc]=size(kdata);\n",
    "% number of frames\n",
    "nt=floor(ntviews/nspokes);\n",
    "% crop the data according to the number of spokes per frame\n",
    "kdata=kdata(:,1:nt*nspokes,:);\n",
    "k=k(:,1:nt*nspokes);\n",
    "w=w(:,1:nt*nspokes);\n",
    "% sort the data into a time-series of undersampled images\n",
    "for ii=1:nt\n",
    "    kdatau(:,:,:,ii)=kdata(:,(ii-1)*nspokes+1:ii*nspokes,:);\n",
    "    ku(:,:,ii)=k(:,(ii-1)*nspokes+1:ii*nspokes);\n",
    "    wu(:,:,ii)=w(:,(ii-1)*nspokes+1:ii*nspokes);\n",
    "end\n",
    "%% normalize smap\n",
    "tmp = sqrt(sum(abs((b1)).^2,3));\n",
    "b1c = div0(b1,tmp);\n",
    "%% prepare for system operator\n",
    "ksp = reshape(ku,[],nt); \n",
    "ksp3(:,1,:) = real(ksp);\n",
    "ksp3(:,2,:) = imag(ksp);\n",
    "om3 = ksp3*2*pi;\n",
    "wi3 = reshape(wu,[],nt);\n",
    "M = size(ksp,1);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.2 Test algorithms\n",
    "\n",
    "**Prepare for PGM algorithms**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "%% prepare for PGM: param\n",
    "param.E = getEnufft(b1c,nt,'M',M,'ksp',ksp3,'om',om3,'wi',wi3,'donufft',1);\n",
    "param.d=permute(reshape(kdatau,[],nc,nt),[1 3 2]);\n",
    "recon_nufft=param.E'*param.d;\n",
    "tscale = 1.3002; %power iteration on E'E\n",
    "param.scaleL = 0.0274/1.099;\n",
    "param.scaleS = 1/1.099;\n",
    "param.lambda_L=0.025;\n",
    "param.lambda_S=0.5*max(abs(recon_nufft(:)))*param.scaleS;\n",
    "param.nite=10;\n",
    "param.tol=0.0025;\n",
    "param.T= getT(nx,ny,nt);\n",
    "param.Xinf = reshape(Xinf.abd,nx*ny,nt);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run ISTA**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 0.1836113, cost: 0.1563813, rank of L: 28\n",
      " ********** PGM: L+S reconstruction **********\n",
      " ite: 1, rank of L: 28, xdiff: 0.1589423, cost: 0.1228233\n",
      " ite: 2, rank of L: 28, xdiff: 0.1390143, cost: 0.0988193\n",
      " ite: 3, rank of L: 28, xdiff: 0.1228113, cost: 0.0814433\n",
      " ite: 4, rank of L: 28, xdiff: 0.1095723, cost: 0.0687143\n",
      " ite: 5, rank of L: 28, xdiff: 0.0987063, cost: 0.0592753\n",
      " ite: 6, rank of L: 28, xdiff: 0.0897413, cost: 0.0521893\n",
      " ite: 7, rank of L: 28, xdiff: 0.0823043, cost: 0.0468013\n",
      " ite: 8, rank of L: 28, xdiff: 0.0760943, cost: 0.0426523\n",
      " ite: 9, rank of L: 28, xdiff: 0.0708683, cost: 0.0394143\n",
      " ite: 10, rank of L: 28, xdiff: 0.0664323, cost: 0.0368533\n",
      "\n",
      "elapsed time: 141.8108\n"
     ]
    }
   ],
   "source": [
    "%% ISTA\n",
    "c = clock();\n",
    "[L_ista,S_ista,xdiff_ista,cost_ista,time_ista,rankL_ista] = PGM(param,'tscale',tscale);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run FISTA**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 0.1836113, cost: 0.1563813, rank of L: 28\n",
      " ********** PGM: L+S reconstruction **********\n",
      " ite: 1, rank of L: 28, xdiff: 0.1711183, cost: 0.1388273\n",
      " ite: 2, rank of L: 28, xdiff: 0.1598423, cost: 0.1239173\n",
      " ite: 3, rank of L: 28, xdiff: 0.1467873, cost: 0.1077923\n",
      " ite: 4, rank of L: 28, xdiff: 0.1327333, cost: 0.0918233\n",
      " ite: 5, rank of L: 28, xdiff: 0.1183843, cost: 0.0770323\n",
      " ite: 6, rank of L: 28, xdiff: 0.1043563, cost: 0.0640723\n",
      " ite: 7, rank of L: 28, xdiff: 0.0911583, cost: 0.0532543\n",
      " ite: 8, rank of L: 28, xdiff: 0.0791733, cost: 0.0446053\n",
      " ite: 9, rank of L: 28, xdiff: 0.0686443, cost: 0.0379483\n",
      " ite: 10, rank of L: 28, xdiff: 0.0596753, cost: 0.0329913\n",
      "\n",
      "elapsed time: 256.7163\n"
     ]
    }
   ],
   "source": [
    "%% FISTA\n",
    "c = clock();\n",
    "[L_fista,S_fista,xdiff_fista,cost_fista,time_fista,rankL_fista] = PGM(param,'fistaL',1,'fistaS',1,'tscale',tscale);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Run POGM**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ite: 0, xdiff: 0.1836113, cost: 0.1563813, rank of L: 28\n",
      " ********** PGM: L+S reconstruction **********\n",
      " ite: 1, rank of L: 28, xdiff: 0.1634303, cost: 0.1285963\n",
      " ite: 2, rank of L: 28, xdiff: 0.1400943, cost: 0.1000613\n",
      " ite: 3, rank of L: 28, xdiff: 0.1161853, cost: 0.0749463\n",
      " ite: 4, rank of L: 28, xdiff: 0.0938363, cost: 0.0553433\n",
      " ite: 5, rank of L: 28, xdiff: 0.0746013, cost: 0.0415433\n",
      " ite: 6, rank of L: 28, xdiff: 0.0593243, cost: 0.0326693\n",
      " ite: 7, rank of L: 26, xdiff: 0.0480713, cost: 0.0273803\n",
      " ite: 8, rank of L: 25, xdiff: 0.0402053, cost: 0.0243883\n",
      " ite: 9, rank of L: 22, xdiff: 0.0345663, cost: 0.0226513\n",
      " ite: 10, rank of L: 21, xdiff: 0.0315303, cost: 0.0219013\n",
      "\n",
      "elapsed time: 140.0721\n"
     ]
    }
   ],
   "source": [
    "%% POGM\n",
    "c = clock();\n",
    "[L_pogm,S_pogm,xdiff_pogm,cost_pogm,time_pogm,rankL_pogm] = PGM(param,'pogmS',1,'pogmL',1,'tscale',tscale);\n",
    "fprintf('\\nelapsed time: %.4f\\n', etime(clock(), c))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Matlab",
   "language": "matlab",
   "name": "matlab"
  },
  "language_info": {
   "codemirror_mode": "octave",
   "file_extension": ".m",
   "help_links": [
    {
     "text": "MetaKernel Magics",
     "url": "https://metakernel.readthedocs.io/en/latest/source/README.html"
    }
   ],
   "mimetype": "text/x-octave",
   "name": "matlab",
   "version": "0.16.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}