Generating Natural-Language Surgical Feedback: From Structured Representation to Domain-Grounded Evaluation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Nasriddinov, Firdavs, Kocielnik, Rafal, Anandkumar, Anima, Hung, Andrew J.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911276285820928
author Nasriddinov, Firdavs
Kocielnik, Rafal
Anandkumar, Anima
Hung, Andrew J.
author_facet Nasriddinov, Firdavs
Kocielnik, Rafal
Anandkumar, Anima
Hung, Andrew J.
contents High-quality intraoperative feedback from a surgical trainer is pivotal for improving trainee performance and long-term skill acquisition. Automating natural, trainer-style feedback promises timely, accessible, and consistent guidance at scale but requires models that understand clinically relevant representations. We present a structure-aware pipeline that learns a surgical action ontology from real trainer-to-trainee transcripts (33 surgeries) and uses it to condition feedback generation. We contribute by (1) mining Instrument-Action-Target (IAT) triplets from real-world feedback text and clustering surface forms into normalized categories, (2) fine-tuning a video-to-IAT model that leverages the surgical procedure and task contexts as well as fine-grained temporal instrument motion, and (3) demonstrating how to effectively use IAT triplet representations to guide GPT-4o in generating clinically grounded, trainer-style feedback. We show that, on Task 1: Video-to-IAT recognition, our context injection and temporal tracking deliver consistent AUC gains (Instrument: 0.67 to 0.74; Action: 0.60 to 0.63; Tissue: 0.74 to 0.79). For Task 2: feedback text generation (rated on a 1-5 fidelity rubric where 1 = opposite/unsafe, 3 = admissible, and 5 = perfect match to a human trainer), GPT-4o from video alone scores 2.17, while IAT conditioning reaches 2.44 (+12.4%), doubling the share of admissible generations with score >= 3 from 21% to 42%. Traditional text-similarity metrics also improve: word error rate decreases by 15-31% and ROUGE (phrase/substring overlap) increases by 9-64%. Grounding generation in explicit IAT structure improves fidelity and yields clinician-verifiable rationales, supporting auditable use in surgical training.
format Preprint
id arxiv_https___arxiv_org_abs_2511_15159
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Generating Natural-Language Surgical Feedback: From Structured Representation to Domain-Grounded Evaluation
Nasriddinov, Firdavs
Kocielnik, Rafal
Anandkumar, Anima
Hung, Andrew J.
Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Machine Learning
68T07, 68T45, 68U10, 68T50, 92C50
I.2; I.2.7; I.2.10; I.5.4; I.4.7; J.3; K.3.1
High-quality intraoperative feedback from a surgical trainer is pivotal for improving trainee performance and long-term skill acquisition. Automating natural, trainer-style feedback promises timely, accessible, and consistent guidance at scale but requires models that understand clinically relevant representations. We present a structure-aware pipeline that learns a surgical action ontology from real trainer-to-trainee transcripts (33 surgeries) and uses it to condition feedback generation. We contribute by (1) mining Instrument-Action-Target (IAT) triplets from real-world feedback text and clustering surface forms into normalized categories, (2) fine-tuning a video-to-IAT model that leverages the surgical procedure and task contexts as well as fine-grained temporal instrument motion, and (3) demonstrating how to effectively use IAT triplet representations to guide GPT-4o in generating clinically grounded, trainer-style feedback. We show that, on Task 1: Video-to-IAT recognition, our context injection and temporal tracking deliver consistent AUC gains (Instrument: 0.67 to 0.74; Action: 0.60 to 0.63; Tissue: 0.74 to 0.79). For Task 2: feedback text generation (rated on a 1-5 fidelity rubric where 1 = opposite/unsafe, 3 = admissible, and 5 = perfect match to a human trainer), GPT-4o from video alone scores 2.17, while IAT conditioning reaches 2.44 (+12.4%), doubling the share of admissible generations with score >= 3 from 21% to 42%. Traditional text-similarity metrics also improve: word error rate decreases by 15-31% and ROUGE (phrase/substring overlap) increases by 9-64%. Grounding generation in explicit IAT structure improves fidelity and yields clinician-verifiable rationales, supporting auditable use in surgical training.
title Generating Natural-Language Surgical Feedback: From Structured Representation to Domain-Grounded Evaluation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Machine Learning
68T07, 68T45, 68U10, 68T50, 92C50
I.2; I.2.7; I.2.10; I.5.4; I.4.7; J.3; K.3.1
url https://arxiv.org/abs/2511.15159