Background Personalized medicine requires the integration and analysis of vast amounts of patient data to realize individualized care. With Surgomics, we aim to facilitate personalized therapy recommendations in surgery by integration of intraoperative surgical data and their analysis with machine learning methods to leverage the potential of this data in analogy to Radiomics and Genomics. Methods We defined Surgomics as the entirety of surgomic features that are process characteristics of a surgical procedure automatically derived from multimodal intraoperative data to quantify processes in the operating room. In a multidisciplinary team we discussed potential data sources like endoscopic videos, vital sign monitoring, medical devices and instruments and respective surgomic features. Subsequently, an online questionnaire was sent to experts from surgery and (computer) science at multiple centers for rating the features’ clinical relevance and technical feasibility. Results In total, 52 surgomic features were identified and assigned to eight feature categories. Based on the expert survey (n = 66 participants) the feature category with the highest clinical relevance as rated by surgeons was “surgical skill and quality of performance” for morbidity and mortality (9.0 ± 1.3 on a numerical rating scale from 1 to 10) as well as for long-term (oncological) outcome (8.2 ± 1.8). The feature category with the highest feasibility to be automatically extracted as rated by (computer) scientists was “Instrument” (8.5 ± 1.7). Among the surgomic features ranked as most relevant in their respective category were “intraoperative adverse events”, “action performed with instruments”, “vital sign monitoring”, and “difficulty of surgery”. Conclusion Surgomics is a promising concept for the analysis of intraoperative data. Surgomics may be used together with preoperative features from clinical data and Radiomics to predict postoperative morbidity, mortality and long-term outcome, as well as to provide tailored feedback for surgeons. Graphical abstract

Aim: We systematically review current clinical applications of artificial intelligence (AI) that use machine learning (ML) methods for decision support in surgical oncology with an emphasis on clinical translation. Methods: MEDLINE, Web of Science, and CENTRAL were searched on 19 January 2021 for a combination of AI and ML-related terms, decision support, and surgical procedures for abdominal malignancies. Data extraction included study characteristics, description of algorithms and their respective purpose, and description of key steps for scientific validation and clinical translation. Results: Out of 8302 articles, 107 studies were included for full-text analysis. Most of the studies were conducted in a retrospective setting (n = 105, 98%), with 45 studies (42%) using data from multiple centers. The most common tumor entities were colorectal cancer (n = 35, 33%), liver cancer (n = 21, 20%), and gastric cancer (n = 17, 16%). The most common prediction task was survival (n = 36, 34%), with artificial neural networks being the most common class of ML algorithms (n = 52, 49%). Key reporting and validation steps included, among others, a complete listing of patient features (n = 95, 89%), training of multiple algorithms (n = 73, 68%), external validation (n = 13, 12%), prospective validation (n = 2, 2%), robustness in terms of cross-validation or resampling (n = 89, 83%), treatment recommendations by ML algorithms (n = 9, 8%), and development of an interface (n = 12, 11%). Conclusion: ML for decision support in surgical oncology is receiving increasing attention with promising results, but robust and prospective clinical validation is mostly lacking. Furthermore, the integration of ML into AI applications is necessary to foster clinical translation.

No abstract