diff --git a/README.md b/README.md
index 5882a19..3f284e5 100644
--- a/README.md
+++ b/README.md
@@ -2,8 +2,9 @@
 
 ##  Evaluation script versioning
 
-#### Current timestamp (YYYY-MM-DD): 2024-02-23
-Update for 2024-02-23 version: Fix `divided by zero exception` Line 405 at `eval_timeline.py`. 
+#### Current timestamp (YYYY-MM-DD): 2024-03-05
+Update for 2024-03-05 version: Added exceptional case handling for situations where a patient doesn’t have gold standard timelines, and the system erroneously predicts timelines. In such cases, the system will now assign a score of 0 for that patient. This is described in detail in [Issue #1](https://github.com/HealthNLPorg/chemoTimelinesEval/issues/1).
+<br>Update for 2024-02-23 version: Fix `divided by zero exception` Line 405 at `eval_timeline.py`. 
 <br>Update for 2024-01-16 version: Initial release.
 
 #### Notice:
diff --git a/eval_timeline.py b/eval_timeline.py
index f420114..bf34b2f 100644
--- a/eval_timeline.py
+++ b/eval_timeline.py
@@ -7,7 +7,7 @@
 
 import dateutil.parser
 
-VERSION = "v20240223"
+VERSION = "v20240305"
 
 logger = logging.getLogger(__name__)
 logger.setLevel(logging.WARNING)
@@ -595,8 +595,14 @@ def macro_average_metrics(
             1.0, 1.0, and 1.0 for the given patient in local precision, recall, and F1, respectively.
         """
         if len(gold_timeline) == 0:
-            true_pos, false_pos, false_neg = [], [], []
-            p, r, f_score = 1, 1, 1
+            if len(pred_timeline) == 0:
+                true_pos, false_pos, false_neg = [], [], []
+                p, r, f_score = 1, 1, 1
+            else:
+                true_pos = []
+                false_pos = pred_timeline
+                false_neg = []
+                p, r, f_score = 0, 0, 0
             local_relations[pred_patient] = 0
         else:
             logger.info(f"pred_patient ID: {pred_patient}")