diff --git a/modules/cold-extraction/ColdDataRetriever.py b/modules/cold-extraction/ColdDataRetriever.py
index 4795c17..f2a8d4e 100644
--- a/modules/cold-extraction/ColdDataRetriever.py
+++ b/modules/cold-extraction/ColdDataRetriever.py
@@ -1,5 +1,5 @@
 import logging
-import os
+import os, glob
 import signal
 import csv
 import time
@@ -13,6 +13,7 @@
 import threading
 import argparse
 import random
+import pandas as pd
 
 from collections import defaultdict
 
@@ -141,7 +142,7 @@ def initialize():
     logging.info("{0}: StoreScp process for the current Niffler extraction is starting now".format(
         datetime.datetime.now()))
 
-    if not storage_folder == "CFIND-ONLY":
+    if not file_path == "CFIND-ONLY":
         subprocess.call("{0}/storescp --accept-unknown --directory {1} --filepath {2} -b {3} > storescp.out &".format(
             DCM4CHE_BIN, storage_folder, file_path, QUERY_AET), shell=True)
 
@@ -211,22 +212,41 @@ def retrieve():
     if number_of_query_attributes > 3 or number_of_query_attributes <= 1:
         # For the cases that extract entirely based on the PatientID - Patient-level extraction.
         if first_attr == "PatientID":
+            temp_folder = "csv/cfind-temp"
+            if not os.path.exists(temp_folder):
+                os.makedirs(temp_folder)
+
             for pid in range(0, length):
                 sleep_for_nightly_mode()
                 patient = firsts[pid]
                 if (not resume) or (resume and (patient not in extracted_ones)):
-                    if storage_folder == "CFIND-ONLY":
-                        inc = random.randint(0,1000000)
-                        subprocess.call("{0}/findscu -c {1} -b {2} -M PatientRoot -m PatientID={3} "
+                    if file_path == "CFIND-ONLY":
+                        if not os.path.exists(temp_folder):
+                            os.makedirs(temp_folder)
+
+                        inc = random.randint(0, 1000000)
+                        subprocess.call("{0}/findscu -c {1} -b {2} -M PatientRoot -m PatientID={3} -r AccessionNumber "
                                         "-r StudyInstanceUID -r StudyDescription -x description.csv.xsl "
-                                        "--out-cat --out-file {4}_{5}.csv --out-dir .".format(
-                            DCM4CHE_BIN, SRC_AET, QUERY_AET,  patient, csv_file, inc), shell=True)
+                                        "--out-cat --out-file {4}/{5}.csv --out-dir .".format(
+                            DCM4CHE_BIN, SRC_AET, QUERY_AET, patient, temp_folder, inc), shell=True)
 
                     else:
                         subprocess.call("{0}/movescu -c {1} -b {2} -M PatientRoot -m PatientID={3} --dest {4}".format(
                             DCM4CHE_BIN, SRC_AET, QUERY_AET, patient, DEST_AET), shell=True)
                     extracted_ones.append(patient)
 
+            if file_path == "CFIND-ONLY":
+                cwd = os.getcwd()
+                os.chdir(temp_folder)
+                all_filenames = [i for i in glob.glob('*.*')]
+                with open(storage_folder + "/cfind-output.csv", 'w') as outfile:
+                    for fname in all_filenames:
+                        with open(fname) as infile:
+                            for line in infile:
+                                outfile.write(line)
+                os.chdir(cwd)
+                shutil.rmtree(temp_folder)
+
         # For the cases that extract based on a single property other than EMPI/PatientID. Goes to study level.
         # "Any" mode. Example: Extractions based on just AccessionNumber of AcquisitionDate.
         else:
diff --git a/modules/cold-extraction/README.md b/modules/cold-extraction/README.md
index d46d403..084ef4f 100644
--- a/modules/cold-extraction/README.md
+++ b/modules/cold-extraction/README.md
@@ -171,7 +171,7 @@ Niffler experimentally supports a CFIND-ONLY mode.
 
 To activate, use the below value,
 ```
-	"StorageFolder": "CFIND-ONLY",
+	"FilePath": "CFIND-ONLY",
 ```
 As of now, this is only supported for the below mode, C-FIND based entirely on PatientID.
 ```
@@ -179,9 +179,8 @@ As of now, this is only supported for the below mode, C-FIND based entirely on P
 	"FirstAttr": "PatientID",
 	"FirstIndex": 0,
 ```
-The output will be a list of CSV files consisting of PatientID, StudyInstanceUI, and StudyDescription.
-
-The CSV outputs will be in the folder of the original CSV file, with the name of CsvFile appended by random numbers.
+The output will be in cfind-output.csv consisting of PatientID, StudyInstanceUID, AccessionNumber, and StudyDescription 
+in the StorageFolder.
 
 ## Troubleshooting
 
diff --git a/modules/cold-extraction/description.csv.xsl b/modules/cold-extraction/description.csv.xsl
index fa0b5ea..986e15b 100644
--- a/modules/cold-extraction/description.csv.xsl
+++ b/modules/cold-extraction/description.csv.xsl
@@ -40,12 +40,16 @@
   <xsl:output method="text"/>
 
   <xsl:template match="/NativeDicomModel">
+    <xsl:text>"</xsl:text>
+    <xsl:apply-templates select="DicomAttribute[@tag='00100020']"/>
+    <xsl:text>"</xsl:text>
+    <xsl:text>,</xsl:text>
     <xsl:text>"</xsl:text>
     <xsl:apply-templates select="DicomAttribute[@tag='0020000D']"/>
     <xsl:text>"</xsl:text>
     <xsl:text>,</xsl:text>
-    <xsl:text>"</xsl:text>    
-    <xsl:apply-templates select="DicomAttribute[@tag='00100020']"/>
+    <xsl:text>"</xsl:text>
+    <xsl:apply-templates select="DicomAttribute[@tag='00080050']"/>
     <xsl:text>"</xsl:text>
     <xsl:text>,</xsl:text>
     <xsl:text>"</xsl:text>