Case study 02

The missing chapter

A case study from the 433-record AI indexing review

What the human summary covered

The human summary covered the patient’s primary complaint thoroughly. It included cholelithiasis (gallstones), cholecystitis workup, a breast rash, hypertension, and seasonal allergies. It was focused and clinically sound for the presenting issue.

What the AI found that the human missed

The AI reconstructed an entire clinical storyline from operative notes and progress notes that the human indexer never summarized:

1. 1
   Choledocholithiasis
   Gallstones that migrated into the common bile duct, a more serious condition than the simple cholelithiasis noted in the summary.
2. 2
   ERCP with endoscopic sphincterotomy
   An interventional procedure to remove the bile duct stones, performed during hospitalization.
3. 3
   Robotic assisted cholecystectomy
   Surgical removal of the gallbladder, following the ERCP.
4. 4
   Postoperative incision infection
   A surgical complication requiring treatment.
5. 5
   Pilonidal cyst infection
   An additional infection documented during the same clinical episode.
6. 6
   Irregular menses
   A gynecological condition documented in clinical notes but absent from the summary.

Why this case matters

This isn’t a missed lab value or a single medication omitted from a list. This is an entire hospitalization, admission, interventional procedure, surgery, surgical complications, and discharge, that does not exist in the human summary. The next clinician reading this summary has no idea this patient underwent abdominal surgery with post-operative complications. If that clinician is planning another procedure, evaluating abdominal symptoms, or assessing surgical risk, they are working from an incomplete picture. The AI didn’t catch one thing the human missed. It caught a whole chapter.

“The AI didn’t catch one thing the human missed. It reconstructed an entire clinical storyline from operative notes the human indexer never summarized.”