Explainable machine-learning predictions for the prevention of hypoxaemia during surgery
By Scott M. Lundberg, Bala G. Nair, M. Vavilala, M. Horibe, Michael J. Eisses, Trevor L. Adams, D. Liston, Daniel K. Low, Shu-Fang Newman, Jerry H. Kim and 1 more. 2018
1,743 citations recorded by the source corpus.
Although anaesthesiologists strive to avoid hypoxaemia during surgery, reliably predicting future intraoperative hypoxaemia is not possible at present. Here, we report the development and testing of a machine-learning-based system that predicts the risk of hypoxaemia and provides explanations of the risk factors in real time during general anaesthesia. The system, which was trained on minute-by-minute data from the electronic medical records of over 50,000 surgeries, improved the performance of anaesthesiologists by providing interpretable hypoxaemia risks and contributing factors. The explanations for the predictions are broadly consistent with the literature and with prior knowledge from anaesthesiologists. Our results suggest that if anaesthesiologists currently anticipate 15% of hypoxaemia events, with the assistance of this system they could anticipate 30%, a large portion of which may benefit from early intervention because they are associated with modifiable factors. The system can help improve the clinical understanding of hypoxaemia risk during anaesthesia care by providing general insights into the exact changes in risk induced by certain characteristics of the patient or procedure. An alert system based on machine learning and trained on surgical data from electronic medical records helps anaesthesiologists prevent hypoxaemia during surgery by providing interpretable real-time predictions.
Subfields: Artificial Intelligence, Cardiology and Cardiovascular Medicine, Health Informatics
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Mohit Bansal
- Percy Liang
- Samuel R. Bowman
- Noah A. Smith
- Luke S. Zettlemoyer
- Jianfeng Gao
- Yejin Choi
- Caiming Xiong
In Cardiology and Cardiovascular Medicine
In Health Informatics
- Eric Horvitz
- Adrian Weller
- Daniel S. Weld
- Ece Kamar
- Marinka Zitnik
- Su-In Lee
- Gregor Stiglic
- Scott M. Lundberg
Citations in the corpus
- Explainable machine learning in deployment (2019, background): “Explanations can come in many forms: from telling patients which symptoms were indicative of a particular diagnosis [35] to helping factory workers analyze inefficiencies in a production pipeline [17].”
- Learning Explainable Models Using Attribution Priors (2019, methodology): “We use the VGG16 architecture introduced by Simonyan and Zisserman [20].”
- Clinically Applicable AI System for Accurate Diagnosis, Quantitative Measurements, and Prognosis of COVID-19 Pneumonia Using Computed Tomography (2020, methodology): “SHAP is a value explainable tool for tree-basedmodels, which could efficiently and exactly compute local explanations and global explanations.”
- Deep Learning for Medical Anomaly Detection – A Survey (2020, neutral): “Image taken from [109]”
- Two-Turn Debate Doesn't Help Humans Answer Hard Reading Comprehension Questions (2022, background): “Previous studies have reported that model-generated explanations can aid humans in some tasks (Cai et al., 2019; Lundberg et al., 2018; Schmidt and Biessmann, 2019; Lai and Tan, 2019), though only when the models are generally able to outperform humans at that task (Bansal et al., 2021).”
- What Clinicians Want: Contextualizing Explainable Machine Learning for Clinical End Use (2019, background): “Although simpler model classes (Lundberg et al., 2018; Wu et al., 2018) are considered more interpetable by users, a general trade-off between model performance and explainability is assumed (Caruana et al., 2015).”
- Interpretability of machine learning‐based prediction models in healthcare (2020, methodology): “For example, SHAP was used in interpretation of predictions for the prevention of hypoxaemia during surgery (Lundberg, et al., 2018), which increased the anaesthesiologists anticipation of hypoxaemia events by 15%, or in the detection of acute intracranial haemorrhage from images of CT scans (Lee,…”
- Explanations Can Reduce Overreliance on AI Systems During Decision-Making (2022, background): “Human-AI performance generally only exceeds human-only performance [8, 26, 30, 55, 65, 78, 97]; stark performance differences between humans and AIs make it such that one would achieve the highest performance”
- Does the Whole Exceed its Parts? The Effect of AI Explanations on Complementary Team Performance (2020, background): “Indeed, at first glance, it appears that researchers have already confirmed the utility of explanations on tasks ranging from medical diagnosis [10, 47] and data annotation [57] to deception detection [36].”
- From explanations to feature selection: assessing SHAP values as feature selection mechanism (2020, positive): “Explanation generated by SHAP [1] visualized with force plot [2] - x axis: data points; y axis: SHAP values/model’s output.”