Perspectives: A surgeon's guide to machine learning

Development and validation of an interpretable Markov-embedded multilabel model for predicting risks of multiple postoperative complications among surgical inpatients: a multicenter prospective cohort study

BACKGROUND

When they encounter various highly related postoperative complications, existing risk evaluation tools that focus on single or any complications are inadequate in clinical practice. This seriously hinders complication management because of the lack of a quantitative basis. An interpretable multilabel model framework that predicts multiple complications simultaneously is urgently needed.

MATERIALS AND METHODS

The authors included 50 325 inpatients from a large multicenter cohort (2014-2017). The authors separated patients from one hospital for external validation and randomly split the remaining patients into training and internal validation sets. A MARKov-EmbeDded (MARKED) multilabel model was proposed, and three models were trained for comparison: binary relevance, a fully connected network (FULLNET), and a deep neural network. Performance was mainly evaluated using the area under the receiver operating characteristic curve (AUC). The authors interpreted the model using Shapley Additive Explanations. Complication-specific risk and risk source inference were provided at the individual level.

RESULTS

There were 26 292, 6574, and 17 459 inpatients in the training, internal validation, and external validation sets, respectively. For the external validation set, MARKED achieved the highest average AUC (0.818, 95% CI: 0.771-0.864) across eight outcomes [compared with binary relevance, 0.799 (0.748-0.849), FULLNET, 0.806 (0.756-0.856), and deep neural network, 0.815 (0.765-0.866)]. Specifically, the AUCs of MARKED were above 0.9 for cardiac complications [0.927 (0.894-0.960)], neurological complications [0.905 (0.870-0.941)], and mortality [0.902 (0.867-0.937)]. Serum albumin, surgical specialties, emergency case, American Society of Anesthesiologists score, age, and sex were the six most important preoperative variables. The interaction between complications contributed more than the preoperative variables, and formed a hierarchical chain of risk factors, mild complications, and severe complications.

CONCLUSION

The authors demonstrated the advantage of MARKED in terms of performance and interpretability. The authors expect that the identification of high-risk patients and the inference of the risk source for specific complications will be valuable for clinical decision-making.

A Bibliometric Analysis of Artificial Intelligence Applications in Spine Care

BACKGROUND

 With the rapid development of science and technology, artificial intelligence (AI) has been widely used in the diagnosis and prognosis of various spine diseases. It has been proved that AI has a broad prospect in accurate diagnosis and treatment of spine disorders.

METHODS

 On May 7, 2022, the Web of Science (WOS) Core Collection database was used to identify the documents on the application of AI in the field of spine care. HistCite and VOSviewer were used for citation analysis and visualization mapping.

RESULTS

 A total of 693 documents were included in the final analysis. The most prolific authors were Karhade A.V. and Schwab J.H. United States was the most productive country. The leading journal was Spine. The most frequently used keyword was spinal. The most prolific institution was Northwestern University in Illinois, USA. Network visualization map showed that United States was the largest network of international cooperation. The keyword "machine learning" had the strongest total link strengths (TLS) and largest number of occurrences. The latest trends suggest that AI for the diagnosis of spine diseases may receive widespread attention in the future.

CONCLUSIONS

 AI has a wide range of application in the field of spine care, and an increasing number of scholars are committed to research on the use of AI in the field of spine care. Bibliometric analysis in the field of AI and spine provides an overall perspective, and the appreciation and research of these influential publications are useful for future research.

Reliability of Postoperative Free Flap Monitoring with a Novel Prediction Model Based on Supervised Machine Learning

BACKGROUND

Postoperative free flap monitoring is a critical part of reconstructive microsurgery. Postoperative clinical assessments rely heavily on specialty-trained staff. Therefore, in regions with limited specialist availability, the feasibility of performing microsurgery is restricted. This study aimed to apply artificial intelligence in postoperative free flap monitoring and validate the ability of machine learning in predicting and differentiating types of postoperative free flap circulation.

METHODS

Postoperative data from 176 patients who received free flap surgery were prospectively collected, including free flap photographs and clinical evaluation measures. Flap circulation outcome variables included normal, arterial insufficiency, and venous insufficiency. The Synthetic Minority Oversampling Technique plus Tomek Links (SMOTE-Tomek) was applied for data balance. Data were divided into 80%:20% for model training and validation. Shapley Additive Explanations were used for prediction interpretations of the model.

RESULTS

Of 805 total included flaps, 555 (69%) were normal, 97 (12%) had arterial insufficiency, and 153 (19%) had venous insufficiency. The most effective prediction model was developed based on random forest, with an accuracy of 98.4%. Temperature and color differences between the flap and the surrounding skin were the most significant contributing factors to predict a vascular compromised flap.

CONCLUSIONS

This study demonstrated the reliability of a machine-learning model in differentiating various types of postoperative flap circulation. This novel technique may reduce the burden of free flap monitoring and encourage the broader use of reconstructive microsurgery in regions with a limited number of staff specialists.

Development and validation of an automated basal cell carcinoma histopathology information extraction system using natural language processing

Introduction

Routinely collected healthcare data are a powerful research resource, but often lack detailed disease-specific information that is collected in clinical free text such as histopathology reports. We aim to use natural Language Processing (NLP) techniques to extract detailed clinical and pathological information from histopathology reports to enrich routinely collected data.

Methods

We used the general architecture for text engineering (GATE) framework to build an NLP information extraction system using rule-based techniques. During validation, we deployed our rule-based NLP pipeline on 200 previously unseen, de-identified and pseudonymised basal cell carcinoma (BCC) histopathological reports from Swansea Bay University Health Board, Wales, UK. The results of our algorithm were compared with gold standard human annotation by two independent and blinded expert clinicians involved in skin cancer care.

Results

We identified 11,224 items of information with a mean precision, recall, and F1 score of 86.0% (95% CI: 75.1-96.9), 84.2% (95% CI: 72.8-96.1), and 84.5% (95% CI: 73.0-95.1), respectively. The difference between clinician annotator F1 scores was 7.9% in comparison with 15.5% between the NLP pipeline and the gold standard corpus. Cohen's Kappa score on annotated tokens was 0.85.

Conclusion

Using an NLP rule-based approach for named entity recognition in BCC, we have been able to develop and validate a pipeline with a potential application in improving the quality of cancer registry data, supporting service planning, and enhancing the quality of routinely collected data for research.

"Beyond MELD" - Emerging strategies and technologies for improving mortality prediction, organ allocation and outcomes in liver transplantation

In this review article, we discuss the model for end-stage liver disease (MELD) score and its dual purpose in general and transplant hepatology. As the landscape of liver disease and transplantation has evolved considerably since the advent of the MELD score, we summarise emerging concepts, methodologies, and technologies that may improve mortality prognostication in the future. Finally, we explore how these novel concepts and technologies may be incorporated into clinical practice.

Artificial Intelligence in Fracture Detection: A Systematic Review and Meta-Analysis

Background Patients with fractures are a common emergency presentation and may be misdiagnosed at radiologic imaging. An increasing number of studies apply artificial intelligence (AI) techniques to fracture detection as an adjunct to clinician diagnosis. Purpose To perform a systematic review and meta-analysis comparing the diagnostic performance in fracture detection between AI and clinicians in peer-reviewed publications and the gray literature (ie, articles published on preprint repositories). Materials and Methods A search of multiple electronic databases between January 2018 and July 2020 (updated June 2021) was performed that included any primary research studies that developed and/or validated AI for the purposes of fracture detection at any imaging modality and excluded studies that evaluated image segmentation algorithms. Meta-analysis with a hierarchical model to calculate pooled sensitivity and specificity was used. Risk of bias was assessed by using a modified Prediction Model Study Risk of Bias Assessment Tool, or PROBAST, checklist. Results Included for analysis were 42 studies, with 115 contingency tables extracted from 32 studies (55 061 images). Thirty-seven studies identified fractures on radiographs and five studies identified fractures on CT images. For internal validation test sets, the pooled sensitivity was 92% (95% CI: 88, 93) for AI and 91% (95% CI: 85, 95) for clinicians, and the pooled specificity was 91% (95% CI: 88, 93) for AI and 92% (95% CI: 89, 92) for clinicians. For external validation test sets, the pooled sensitivity was 91% (95% CI: 84, 95) for AI and 94% (95% CI: 90, 96) for clinicians, and the pooled specificity was 91% (95% CI: 81, 95) for AI and 94% (95% CI: 91, 95) for clinicians. There were no statistically significant differences between clinician and AI performance. There were 22 of 42 (52%) studies that were judged to have high risk of bias. Meta-regression identified multiple sources of heterogeneity in the data, including risk of bias and fracture type. Conclusion Artificial intelligence (AI) and clinicians had comparable reported diagnostic performance in fracture detection, suggesting that AI technology holds promise as a diagnostic adjunct in future clinical practice. Clinical trial registration no. CRD42020186641 © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Cohen and McInnes in this issue.