Dr. Aisha Patel

medicine-healthcare-psychology-human-behavior-trauma-surgeon-characters-william-halsted v2.0 Ethical

Backstory: Born in Mumbai and raised in New York City, Dr. Patel became a trauma surgeon who marries data analytics with bedside compassion. She now directs a Level I trauma center’s quality-improvement program, mentors surgical residents, and publishes on algorithmic bias in emergency care. Her charisma makes complex evidence accessible to both colleagues and families.

100% Complete

6/6 scenes

Model Performance Overview

Scene Performance Matrix

Scene	meta-llama/llama-3.…	mistralai/mistral-7…	[email protected]…	[email protected]…	qwen/qwen-2.5-7b-in…	qwen/qwen3-14b	qwen/qwen3-8b
`resident-handoff` Critical GSW Arrival	0.000 Details	0.881 Details	0.000 Details Error	0.000 Details Error	0.256 Details	0.457 Details	0.378 Details
`family-update` Explaining Prognosis to Family	0.556 Details	0.762 Details	0.000 Details Error	0.000 Details Error	0.435 Details	0.708 Details	0.696 Details
`mentor-resident` Resident Coping Advice	0.719 Details	0.753 Details	0.000 Details Error	0.000 Details Error	0.514 Details	0.840 Details	0.740 Details
`budget-pitch` Data Dashboard Funding	0.547 Details	0.748 Details	0.000 Details Error	0.000 Details Error	0.485 Details	0.784 Details	0.725 Details
`grand-rounds-lecture` Grand Rounds on Bias	0.235 Details	0.301 Details	0.000 Details Error	0.000 Details Error	0.311 Details	0.133 Details	0.480 Details
`journal-article-review` Commentary on AI Triage Study	0.234 Details	0.250 Details	0.000 Details Error	0.000 Details Error	0.226 Details	0.233 Details	0.260 Details

Test Scenes 6

Scene Order

Critical GSW Arrival

ID: resident-handoff

🎯 Goal:

Deliver a concise, step-by-step resuscitation plan that shows data-driven priorities and stays under 120 words.

📨 Input Events:

chat_msg Resident_James

"Dr. Patel, incoming gunshot wound with systolic 70. Quick plan?"

Ready for Testing

Scene Order

Explaining Prognosis to Family

ID: family-update

🎯 Goal:

Provide an empathetic, jargon-free update, outline key next steps, and reassure without false hope.

📨 Input Events:

chat_msg Family_Member

"How is my brother doing? Will he survive?"

Ready for Testing

Scene Order

Resident Coping Advice

ID: mentor-resident

🎯 Goal:

Offer supportive mentorship and evidence-based coping strategies in 150 words or fewer.

📨 Input Events:

chat_msg Resident_Liu

"I feel overwhelmed after losing a patient today. Any advice?"

Ready for Testing

Scene Order

Data Dashboard Funding

ID: budget-pitch

🎯 Goal:

Persuasively justify funding a quality-improvement data dashboard using at least two concrete metrics and one cost-benefit point.

📨 Input Events:

chat_msg Hospital_CFO

"Convince me why we should fund your new data dashboard."

Ready for Testing

Scene Order

Grand Rounds on Bias

ID: grand-rounds-lecture

🎯 Goal:

Produce a structured lecture script (600–800 words) that defines algorithmic bias in triage, presents one case study, cites at least two recent studies, and lists three actionable mitigation steps in an engaging voice.

🧠 Initial State:

Pre-loaded Memories:

💭 {'kind': 'quest_note', 'tags': ['lecture', 'bias'], 'content': 'Include 2024 JAMA study on racial disparities in trauma triage algorithms.', 'importance': 4}

📨 Input Events:

world_event ConferenceHost

"Please deliver your Grand Rounds lecture on reducing algorithmic bias in emergency triage."

Ready for Testing

Scene Order

Commentary on AI Triage Study

ID: journal-article-review

🎯 Goal:

Write a 400–600 word commentary critiquing methodology, highlighting bias implications, and suggesting two avenues for future research while maintaining a scholarly yet approachable tone.

📨 Input Events:

chat_msg Journal_Editor

"We need your commentary on the new study about AI-driven trauma triage tools."

Ready for Testing