Artificial intelligence for real-time surgical phase recognition in minimal invasive inguinal hernia repair: a systematic review on behalf of TROGSS - the robotic global surgical society
Bibliographic record
Abstract
Introduction: Artificial intelligence (AI) integration into surgical practice has advanced intraoperative precision, complication prediction, and procedural efficiency. While AI has demonstrated advancements in colorectal, cardiac, and other laparoscopic procedures, its application in inguinal hernia repair (IHR), one of the most commonly performed surgeries, remains underexplored. AI models demonstrate potential in real-time recognition of surgical phases, anatomical structures, and instruments, particularly in transabdominal preperitoneal (TAPP), total extraperitoneal (TEP), and robotic inguinal hernia repair (RIHR). This systematic review evaluates the accuracy, applicability, and clinical impact of AI-based systems in real-time surgical phase recognition during IHR. Methods: Following PRISMA 2020 guidelines and PROSPERO registration (CRD42024621178), a systematic search of PubMed, Scopus, Web of Science, Embase, Cochrane Library, and ScienceDirect was conducted on November 12, 2024. Studies utilizing AI models for real-time video-based surgical phase recognition in minimally invasive IHR (TAPP, TEP, and RIHR) were included. The screening process, data extraction task, and quality assessment using NOS (Newcastle-Ottawa Scale) were performed by three independent reviewers. Primary outcomes were AI performance metrics (accuracy, F1-score, precision, recall, and latency), and secondary outcomes included clinical phase recognition performance. Results: Out of 903 records, six studies (2022-2024) were included, involving laparoscopic (n = 4) and robotic-assisted (n = 2) IHR from the United States (n = 2), France (n = 2), and Greece (n = 1). A total of 774 videos (25-619 per study) underwent pre-processing (frame extraction or down-sampling). Annotation tools included CVAT, SuperAnnotate, and manual labeling. AI models (VTN, DETR, ResNet-50, YOLOv8) demonstrated accuracy between 74% and > 87%, with YOLOv8 achieving the highest F1-score (82%). Risk of bias was moderate to high, with Fleiss’ kappa for inter-rater agreement at 0.82 (selection) and 0.49 (comparability). Conclusion: AI and ML models demonstrate significant potential in achieving real-time surgical phase recognition during minimally invasive IHR. Despite promising accuracies, challenges such as heterogeneity in model performance, reliance on annotated datasets, and the need for real-time validation persist. Standardized benchmarks, multicenter studies, and hardware advancements will be essential to fully integrate AI into surgical workflows, improving surgical training, technical performance, and patient outcomes.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.014 | 0.059 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.005 | 0.006 |
| Bibliometrics | 0.011 | 0.010 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.006 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".