Deep Learning-Assisted Detection of Glaucoma Progression in Spectral-Domain OCT.

PURPOSE: To develop and validate a deep learning (DL) model for detection of glaucoma progression using spectral-domain (SD)-OCT measurements of retinal nerve fiber layer (RNFL) thickness. DESIGN: Retrospective cohort study. PARTICIPANTS: A total of 14 034 SD-OCT scans from 816 eyes from 462 individuals. METHODS: A DL convolutional neural network was trained to assess SD-OCT RNFL thickness measurements of 2 visits (a baseline and a follow-up visit) along with time between visits to predict the probability of glaucoma progression. The ground truth was defined by consensus from subjective grading by glaucoma specialists. Diagnostic performance was summarized by the area under the receiver operator characteristic curve (AUC), sensitivity, and specificity, and was compared with conventional trend-based analyses of change. Interval likelihood ratios were calculated to determine the impact of DL model results in changing the post-test probability of progression. MAIN OUTCOME MEASURES: The AUC, sensitivity, and specificity of the DL model. RESULTS: The DL model had an AUC of 0.938 (95% confidence interval [CI], 0.921-0.955), with sensitivity of 87.3% (95% CI, 83.6%-91.6%) and specificity of 86.4% (95% CI, 79.9%-89.6%). When matched for the same specificity, the DL model significantly outperformed trend-based analyses. Likelihood ratios for the DL model were associated with large changes in the probability of progression in the vast majority of SD-OCT tests. CONCLUSIONS: A DL model was able to assess the probability of glaucomatous structural progression from SD-OCT RNFL thickness measurements. The model agreed well with expert judgments and outperformed conventional trend-based analyses of change, while also providing indication of the likely locations of change. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found after the references.

Performance of a Machine Learning Algorithm Using Electronic Health Record Data to Predict Postoperative Complications and Report on a Mobile Platform.

Importance: Predicting postoperative complications has the potential to inform shared decisions regarding the appropriateness of surgical procedures, targeted risk-reduction strategies, and postoperative resource use. Realizing these advantages requires that accurate real-time predictions be integrated with clinical and digital workflows; artificial intelligence predictive analytic platforms using automated electronic health record (EHR) data inputs offer an intriguing possibility for achieving this, but there is a lack of high-level evidence from prospective studies supporting their use. Objective: To examine whether the MySurgeryRisk artificial intelligence system has stable predictive performance between development and prospective validation phases and whether it is feasible to provide automated outputs directly to surgeons' mobile devices. Design, Setting, and Participants: In this prognostic study, the platform used automated EHR data inputs and machine learning algorithms to predict postoperative complications and provide predictions to surgeons, previously through a web portal and currently through a mobile device application. All patients 18 years or older who were admitted for any type of inpatient surgical procedure (74 417 total procedures involving 58 236 patients) between June 1, 2014, and September 20, 2020, were included. Models were developed using retrospective data from 52 117 inpatient surgical procedures performed between June 1, 2014, and November 27, 2018. Validation was performed using data from 22 300 inpatient surgical procedures collected prospectively from November 28, 2018, to September 20, 2020. Main Outcomes and Measures: Algorithms for generalized additive models and random forest models were developed and validated using real-time EHR data. Model predictive performance was evaluated primarily using area under the receiver operating characteristic curve (AUROC) values. Results: Among 58 236 total adult patients who received 74 417 major inpatient surgical procedures, the mean (SD) age was 57 (17) years; 29 226 patients (50.2%) were male. Results reported in this article focus primarily on the validation cohort. The validation cohort included 22 300 inpatient surgical procedures involving 19 132 patients (mean [SD] age, 58 [17] years; 9672 [50.6%] male). A total of 2765 patients (14.5%) were Black or African American, 14 777 (77.2%) were White, 1235 (6.5%) were of other races (including American Indian or Alaska Native, Asian, Native Hawaiian or Pacific Islander, and multiracial), and 355 (1.9%) were of unknown race because of missing data; 979 patients (5.1%) were Hispanic, 17 663 (92.3%) were non-Hispanic, and 490 (2.6%) were of unknown ethnicity because of missing data. A greater number of input features was associated with stable or improved model performance. For example, the random forest model trained with 135 input features had the highest AUROC values for predicting acute kidney injury (0.82; 95% CI, 0.82-0.83); cardiovascular complications (0.81; 95% CI, 0.81-0.82); neurological complications, including delirium (0.87; 95% CI, 0.87-0.88); prolonged intensive care unit stay (0.89; 95% CI, 0.88-0.89); prolonged mechanical ventilation (0.91; 95% CI, 0.90-0.91); sepsis (0.86; 95% CI, 0.85-0.87); venous thromboembolism (0.82; 95% CI, 0.81-0.83); wound complications (0.78; 95% CI, 0.78-0.79); 30-day mortality (0.84; 95% CI, 0.82-0.86); and 90-day mortality (0.84; 95% CI, 0.82-0.85), with accuracy similar to surgeons' predictions. Compared with the original web portal, the mobile device application allowed efficient fingerprint login access and loaded data approximately 10 times faster. The application output displayed patient information, risk of postoperative complications, top 3 risk factors for each complication, and patterns of complications for individual surgeons compared with their colleagues. Conclusions and Relevance: In this study, automated real-time predictions of postoperative complications with mobile device outputs had good performance in clinical settings with prospective validation, matching surgeons' predictive accuracy.

Physiologic signatures within six hours of hospitalization identify acute illness phenotypes.

During the early stages of hospital admission, clinicians use limited information to make decisions as patient acuity evolves. We hypothesized that clustering analysis of vital signs measured within six hours of hospital admission would reveal distinct patient phenotypes with unique pathophysiological signatures and clinical outcomes. We created a longitudinal electronic health record dataset for 75,762 adult patient admissions to a tertiary care center in 2014-2016 lasting six hours or longer. Physiotypes were derived via unsupervised machine learning in a training cohort of 41,502 patients applying consensus k-means clustering to six vital signs measured within six hours of admission. Reproducibility and correlation with clinical biomarkers and outcomes were assessed in validation cohort of 17,415 patients and testing cohort of 16,845 patients. Training, validation, and testing cohorts had similar age (54-55 years) and sex (55% female), distributions. There were four distinct clusters. Physiotype A had physiologic signals consistent with early vasoplegia, hypothermia, and low-grade inflammation and favorable short-and long-term clinical outcomes despite early, severe illness. Physiotype B exhibited early tachycardia, tachypnea, and hypoxemia followed by the highest incidence of prolonged respiratory insufficiency, sepsis, acute kidney injury, and short- and long-term mortality. Physiotype C had minimal early physiological derangement and favorable clinical outcomes. Physiotype D had the greatest prevalence of chronic cardiovascular and kidney disease, presented with severely elevated blood pressure, and had good short-term outcomes but suffered increased 3-year mortality. Comparing sequential organ failure assessment (SOFA) scores across physiotypes demonstrated that clustering did not simply recapitulate previously established acuity assessments. In a heterogeneous cohort of hospitalized patients, unsupervised machine learning techniques applied to routine, early vital sign data identified physiotypes with unique disease categories and distinct clinical outcomes. This approach has the potential to augment understanding of pathophysiology by distilling thousands of disease states into a few physiological signatures.

RetiNerveNet: using recursive deep learning to estimate pointwise 24-2 visual field data based on retinal structure.

Glaucoma is the leading cause of irreversible blindness in the world, affecting over 70 million people. The cumbersome Standard Automated Perimetry (SAP) test is most frequently used to detect visual loss due to glaucoma. Due to the SAP test's innate difficulty and its high test-retest variability, we propose the RetiNerveNet, a deep convolutional recursive neural network for obtaining estimates of the SAP visual field. RetiNerveNet uses information from the more objective Spectral-Domain Optical Coherence Tomography (SDOCT). RetiNerveNet attempts to trace-back the arcuate convergence of the retinal nerve fibers, starting from the Retinal Nerve Fiber Layer (RNFL) thickness around the optic disc, to estimate individual age-corrected 24-2 SAP values. Recursive passes through the proposed network sequentially yield estimates of the visual locations progressively farther from the optic disc. While all the methods used for our experiments exhibit lower performance for the advanced disease group (possibly due to the "floor effect" for the SDOCT test), the proposed network is observed to be more accurate than all the baselines for estimating the individual visual field values. We further augment the proposed network to additionally predict the SAP Mean Deviation values and also facilitate the assignment of higher weightage to the underrepresented groups in the data. We then study the resulting performance trade-offs of the RetiNerveNet on the early, moderate and severe disease groups.

Optimizing predictive strategies for acute kidney injury after major vascular surgery.

BACKGROUND: Postoperative acute kidney injury is common after major vascular surgery and is associated with increased morbidity, mortality, and cost. High-performance risk stratification using a machine learning model can inform strategies that mitigate harm and optimize resource use. It is hypothesized that incorporating intraoperative data would improve machine learning model accuracy, discrimination, and precision in predicting acute kidney injury among patients undergoing major vascular surgery. METHODS: A single-center retrospective cohort of 1,531 adult patients who underwent nonemergency major vascular surgery, including open aortic, endovascular aortic, and lower extremity bypass procedures, was evaluated. The validated, automated MySurgeryRisk analytics platform used electronic health record data to forecast patient-level probabilistic risk scores for postoperative acute kidney injury using random forest models with preoperative data alone and perioperative data (preoperative plus intraoperative). The MySurgeryRisk predictions were compared with each other as well as with the American Society of Anesthesiologists physical status classification. RESULTS: Machine learning models using perioperative data had greater accuracy, discrimination, and precision than models using either preoperative data alone or the American Society of Anesthesiologists physical status classification (accuracy: 0.70 vs 0.64 vs 0.62, area under the receiver operating characteristics curve: 0.77 vs 0.68 vs 0.61, area under the precision-recall curve: 0.70 vs 0.58 vs 0.48). CONCLUSION: In predicting acute kidney injury after major vascular surgery, machine learning approaches that incorporate dynamic intraoperative data had greater accuracy, discrimination, and precision than models using either preoperative data alone or the American Society of Anesthesiologists physical status classification. Machine learning methods have the potential for real-time identification of high-risk patients who may benefit from personalized risk-reduction strategies.

Reinforcement learning in surgery.

Patients and physicians make essential decisions regarding diagnostic and therapeutic interventions. These actions should be performed or deferred under time constraints and uncertainty regarding patients' diagnoses and predicted response to treatment. This may lead to cognitive and judgment errors. Reinforcement learning is a subfield of machine learning that identifies a sequence of actions to increase the probability of achieving a predetermined goal. Reinforcement learning has the potential to assist in surgical decision making by recommending actions at predefined intervals and its ability to utilize complex input data, including text, image, and temporal data, in the decision-making process. The algorithm mimics a human trial-and-error learning process to calculate optimum recommendation policies. The article provides insight regarding challenges in the development and application of reinforcement learning in the medical field, with an emphasis on surgical decision making. The review focuses on challenges in formulating reward function describing the ultimate goal and determination of patient states derived from electronic health records, along with the lack of resources to simulate the potential benefits of suggested actions in response to changing physiological states during and after surgery. Although clinical implementation would require secure, interoperable, livestreaming electronic health record data for use by virtual model, development and validation of personalized reinforcement learning models in surgery can contribute to improving care by helping patients and clinicians make better decisions.

Fuzzy Clustering to Identify Clusters at Different Levels of Fuzziness: An Evolutionary Multiobjective Optimization Approach.

Fuzzy clustering methods identify naturally occurring clusters in a dataset, where the extent to which different clusters are overlapped can differ. Most methods have a parameter to fix the level of fuzziness. However, the appropriate level of fuzziness depends on the application at hand. This paper presents an entropy c -means (ECM), a method of fuzzy clustering that simultaneously optimizes two contradictory objective functions, resulting in the creation of fuzzy clusters with different levels of fuzziness. This allows ECM to identify clusters with different degrees of overlap. ECM optimizes the two objective functions using two multiobjective optimization methods, nondominated sorting genetic algorithm II (NSGA-II) and multiobjective evolutionary algorithm based on decomposition (MOEA/D). We also propose a method to select a suitable tradeoff clustering from the Pareto front. Experiments on challenging synthetic datasets as well as real-world datasets show that ECM leads to better cluster detection compared to the conventional fuzzy clustering methods as well as previously used multiobjective methods for fuzzy clustering.

Artificial Intelligence Mapping of Structure to Function in Glaucoma.

Purpose: To develop an artificial intelligence (AI)-based structure-function (SF) map relating retinal nerve fiber layer (RNFL) damage on spectral domain optical coherence tomography (SDOCT) to functional loss on standard automated perimetry (SAP). Methods: The study included 26,499 pairs of SAP and SDOCT from 15,173 eyes of 8878 patients with glaucoma or suspected of having the disease extracted from the Duke Glaucoma Registry. The data set was randomly divided at the patient level in training and test sets. A convolutional neural network (CNN) was initially trained and validated to predict the 52 sensitivity threshold points of the 24-2 SAP from the 768 RNFL thickness points of the SDOCT peripapillary scan. Simulated localized RNFL defects of varied locations and depths were created by modifying the normal average peripapillary RNFL profile. The simulated profiles were then fed to the previously trained CNN, and the topographic SF relationships between structural defects and SAP functional losses were investigated. Results: The CNN predictions had an average correlation coefficient of 0.60 (P < 0.001) with the measured values from SAP and a mean absolute error of 4.25 dB. Simulated RNFL defects led to well-defined arcuate or paracentral visual field losses in the opposite hemifield, which varied according to the location and depth of the simulations. Conclusions: A CNN was capable of predicting SAP sensitivity thresholds from SDOCT RNFL thickness measurements and generate an SF map from simulated defects. Translational Relevance: AI-based SF map improves the understanding of how SDOCT losses translate into detectable SAP damage.

Added Value of Intraoperative Data for Predicting Postoperative Complications: The MySurgeryRisk PostOp Extension.

BACKGROUND: Models that predict postoperative complications often ignore important intraoperative events and physiological changes. This study tested the hypothesis that accuracy, discrimination, and precision in predicting postoperative complications would improve when using both preoperative and intraoperative data input data compared with preoperative data alone. METHODS: This retrospective cohort analysis included 43,943 adults undergoing 52,529 inpatient surgeries at a single institution during a 5-y period. Random forest machine learning models in the validated MySurgeryRisk platform made patient-level predictions for seven postoperative complications and mortality occurring during hospital admission using electronic health record data and patient neighborhood characteristics. For each outcome, one model trained with preoperative data alone; one model trained with both preoperative and intraoperative data. Models were compared by accuracy, discrimination (expressed as area under the receiver operating characteristic curve), precision (expressed as area under the precision-recall curve), and reclassification indices. RESULTS: Machine learning models incorporating both preoperative and intraoperative data had greater accuracy, discrimination, and precision than models using preoperative data alone for predicting all seven postoperative complications (intensive care unit length of stay >48 h, mechanical ventilation >48 h, neurologic complications including delirium, cardiovascular complications, acute kidney injury, venous thromboembolism, and wound complications), and in-hospital mortality (accuracy: 88% versus 77%; area under the receiver operating characteristic curve: 0.93 versus 0.87; area under the precision-recall curve: 0.21 versus 0.15). Overall reclassification improvement was 2.4%-10.0% for complications and 11.2% for in-hospital mortality. CONCLUSIONS: Incorporating both preoperative and intraoperative data significantly increased the accuracy, discrimination, and precision of machine learning models predicting postoperative complications and mortality.

Multiobjective Support Vector Machines: Handling Class Imbalance With Pareto Optimality.

Support vector machines (SVMs) seek to optimize three distinct objectives: maximization of margin, minimization of regularization from the positive class, and minimization of regularization from the negative class. The right choice of weightage for each of these objectives is critical to the quality of the classifier learned, especially in case of the class imbalanced data sets. Therefore, costly parameter tuning has to be undertaken to find a set of suitable relative weights. In this brief, we propose to train SVMs, on two-class as well as multiclass data sets, in a multiobjective optimization framework called radial boundary intersection to overcome this shortcoming. The experimental results suggest that the radial boundary intersection-based scheme is indeed effective in finding the best tradeoff among the objectives compared with parameter-tuning schemes.

Adaptive Learning-Based -Nearest Neighbor Classifiers With Resilience to Class Imbalance.

The classification accuracy of a -nearest neighbor ( NN) classifier is largely dependent on the choice of the number of nearest neighbors denoted by . However, given a data set, it is a tedious task to optimize the performance of NN by tuning . Moreover, the performance of NN degrades in the presence of class imbalance, a situation characterized by disparate representation from different classes. We aim to address both the issues in this paper and propose a variant of NN called the Adaptive NN (Ada- NN). The Ada- NN classifier uses the density and distribution of the neighborhood of a test point and learns a suitable point-specific for it with the help of artificial neural networks. We further improve our proposal by replacing the neural network with a heuristic learning method guided by an indicator of the local density of a test point and using information about its neighboring training points. The proposed heuristic learning algorithm preserves the simplicity of NN without incurring serious computational burden. We call this method Ada- NN2. Ada- NN and Ada- NN2 perform very competitive when compared with NN, five of NN's state-of-the-art variants, and other popular classifiers. Furthermore, we propose a class-based global weighting scheme (Global Imbalance Handling Scheme or GIHS) to compensate for the effect of class imbalance. We perform extensive experiments on a wide variety of data sets to establish the improvement shown by Ada- NN and Ada- NN2 using the proposed GIHS, when compared with NN, and its 12 variants specifically tailored for imbalanced classification.

Near-Bayesian Support Vector Machines for imbalanced data classification with equal or unequal misclassification costs.

Support Vector Machines (SVMs) form a family of popular classifier algorithms originally developed to solve two-class classification problems. However, SVMs are likely to perform poorly in situations with data imbalance between the classes, particularly when the target class is under-represented. This paper proposes a Near-Bayesian Support Vector Machine (NBSVM) for such imbalanced classification problems, by combining the philosophies of decision boundary shift and unequal regularization costs. Based on certain assumptions which hold true for most real-world datasets, we use the fractions of representation from each of the classes, to achieve the boundary shift as well as the asymmetric regularization costs. The proposed approach is extended to the multi-class scenario and also adapted for cases with unequal misclassification costs for the different classes. Extensive comparison with standard SVM and some state-of-the-art methods is furnished as a proof of the ability of the proposed approach to perform competitively on imbalanced datasets. A modified Sequential Minimal Optimization (SMO) algorithm is also presented to solve the NBSVM optimization problem in a computationally efficient manner.