Identification of predictive patient characteristics for assessing the probability of COVID-19 in-hospital mortality.

As the world emerges from the COVID-19 pandemic, there is an urgent need to understand patient factors that may be used to predict the occurrence of severe cases and patient mortality. Approximately 20% of SARS-CoV-2 infections lead to acute respiratory distress syndrome caused by the harmful actions of inflammatory mediators. Patients with severe COVID-19 are often afflicted with neurologic symptoms, and individuals with pre-existing neurodegenerative disease have an increased risk of severe COVID-19. Although collectively, these observations point to a bidirectional relationship between severe COVID-19 and neurologic disorders, little is known about the underlying mechanisms. Here, we analyzed the electronic health records of 471 patients with severe COVID-19 to identify clinical characteristics most predictive of mortality. Feature discovery was conducted by training a regularized logistic regression classifier that serves as a machine-learning model with an embedded feature selection capability. SHAP analysis using the trained classifier revealed that a small ensemble of readily observable clinical features, including characteristics associated with cognitive impairment, could predict in-hospital mortality with an accuracy greater than 0.85 (expressed as the area under the ROC curve of the classifier). These findings have important implications for the prioritization of clinical measures used to identify patients with COVID-19 (and, potentially, other forms of acute respiratory distress syndrome) having an elevated risk of death.

Deep-Learning-Based High-Intensity Focused Ultrasound Lesion Segmentation in Multi-Wavelength Photoacoustic Imaging.

Photoacoustic (PA) imaging can be used to monitor high-intensity focused ultrasound (HIFU) therapies because ablation changes the optical absorption spectrum of the tissue, and this change can be detected with PA imaging. Multi-wavelength photoacoustic (MWPA) imaging makes this change easier to detect by repeating PA imaging at multiple optical wavelengths and sampling the optical absorption spectrum more thoroughly. Real-time pixel-wise classification in MWPA imaging can assist clinicians in monitoring HIFU lesion formation and will be a crucial milestone towards full HIFU therapy automation based on artificial intelligence. In this paper, we present a deep-learning-based approach to segment HIFU lesions in MWPA images. Ex vivo bovine tissue is ablated with HIFU and imaged via MWPA imaging. The acquired MWPA images are then used to train and test a convolutional neural network (CNN) for lesion segmentation. Traditional machine learning algorithms are also trained and tested to compare with the CNN, and the results show that the performance of the CNN significantly exceeds traditional machine learning algorithms. Feature selection is conducted to reduce the number of wavelengths to facilitate real-time implementation while retaining good segmentation performance. This study demonstrates the feasibility and high performance of the deep-learning-based lesion segmentation method in MWPA imaging to monitor HIFU lesion formation and the potential to implement this method in real time.

A machine learning model for orthodontic extraction/non-extraction decision in a racially and ethnically diverse patient population.

INTRODUCTION: The purpose of the present study was to create a machine learning (ML) algorithm with the ability to predict the extraction/non-extraction decision in a racially and ethnically diverse sample. METHODS: Data was gathered from the records of 393 patients (200 non-extraction and 193 extraction) from a racially and ethnically diverse population. Four ML models (logistic regression [LR], random forest [RF], support vector machine [SVM], and neural network [NN]) were trained on a training set (70% of samples) and then tested on the remaining samples (30%). The accuracy and precision of the ML model predictions were calculated using the area under the curve (AUC) of the receiver operating characteristics (ROC) curve. The proportion of correct extraction/non-extraction decisions was also calculated. RESULTS: The LR, SVM, and NN models performed best, with an AUC of the ROC of 91.0%, 92.5%, and 92.3%, respectively. The overall proportion of correct decisions was 82%, 76%, 83%, and 81% for the LR, RF, SVM, and NN models, respectively. The features found to be most helpful to the ML algorithms in making their decisions were maxillary crowding/spacing, L1-NB (mm), U1-NA (mm), PFH:AFH, and SN-MP(̊), although many other features contributed significantly. CONCLUSIONS: ML models can predict the extraction decision in a racially and ethnically diverse patient population with a high degree of accuracy and precision. Crowding, sagittal, and vertical characteristics all featured prominently in the hierarchy of components most influential to the ML decision-making process.

A novel statistical methodology for quantifying the spatial arrangements of axons in peripheral nerves.

A thorough understanding of the neuroanatomy of peripheral nerves is required for a better insight into their function and the development of neuromodulation tools and strategies. In biophysical modeling, it is commonly assumed that the complex spatial arrangement of myelinated and unmyelinated axons in peripheral nerves is random, however, in reality the axonal organization is inhomogeneous and anisotropic. Present quantitative neuroanatomy methods analyze peripheral nerves in terms of the number of axons and the morphometric characteristics of the axons, such as area and diameter. In this study, we employed spatial statistics and point process models to describe the spatial arrangement of axons and Sinkhorn distances to compute the similarities between these arrangements (in terms of first- and second-order statistics) in various vagus and pelvic nerve cross-sections. We utilized high-resolution transmission electron microscopy (TEM) images that have been segmented using a custom-built high-throughput deep learning system based on a highly modified U-Net architecture. Our findings show a novel and innovative approach to quantifying similarities between spatial point patterns using metrics derived from the solution to the optimal transport problem. We also present a generalizable pipeline for quantitative analysis of peripheral nerve architecture. Our data demonstrate differences between male- and female-originating samples and similarities between the pelvic and abdominal vagus nerves.

Can we predict orthodontic extraction patterns by using machine learning?

OBJECTIVE: To investigate the utility of machine learning (ML) in accurately predicting orthodontic extraction patterns in a heterogeneous population. MATERIALS AND METHODS: The material of this retrospective study consisted of records of 366 patients treated with orthodontic extractions. The dataset was randomly split into training (70%) and test sets (30%) and was stratified according to race/ethnicity and gender. Fifty-five cephalometric and demographic input data were used to train and test multiple ML algorithms. The extraction patterns were labelled according to the previous treatment plan. Random Forest (RF), Logistic Regression (LR), and Support Vector Machine (SVM) algorithms were used to predict the patient's extraction patterns. RESULTS: The highest class accuracy percentages were obtained for the upper and lower 1st premolars (U/L4s) (RF: 81.63%, LR: 63.27%, SVM: 63.27%) and upper 1st premolars only (U4s) extraction patterns (RF: 61.11%, LR: 72.22%, SVM: 72.22%). However, all methods revealed low class accuracy rates (<50%) for the upper 1st and lower 2nd premolars (U4/L5s), upper 2nd and lower 1st premolars (U5/L4s), and upper and lower 2nd premolars (U/L5s) extraction patterns. For the overall accuracy, RF yielded the highest percentage with 54.55%, followed by SVM with 52.73% and LR with 49.09%. CONCLUSION: All tested supervised ML techniques yielded good accuracy in predicting U/L4s and U4s extraction patterns. However, they predicted poorly for the U4/L5s, U5/L4s, and U/L5s extraction patterns. Molar relationship, mandibular crowding, and overjet were found to be the most predictive indicators for determining extraction patterns.

High-throughput segmentation of unmyelinated axons by deep learning.

Axonal characterizations of connectomes in healthy and disease phenotypes are surprisingly incomplete and biased because unmyelinated axons, the most prevalent type of fibers in the nervous system, have largely been ignored as their quantitative assessment quickly becomes unmanageable as the number of axons increases. Herein, we introduce the first prototype of a high-throughput processing pipeline for automated segmentation of unmyelinated fibers. Our team has used transmission electron microscopy images of vagus and pelvic nerves in rats. All unmyelinated axons in these images are individually annotated and used as labeled data to train and validate a deep instance segmentation network. We investigate the effect of different training strategies on the overall segmentation accuracy of the network. We extensively validate the segmentation algorithm as a stand-alone segmentation tool as well as in an expert-in-the-loop hybrid segmentation setting with preliminary, albeit remarkably encouraging results. Our algorithm achieves an instance-level [Formula: see text] score of between 0.7 and 0.9 on various test images in the stand-alone mode and reduces expert annotation labor by 80% in the hybrid setting. We hope that this new high-throughput segmentation pipeline will enable quick and accurate characterization of unmyelinated fibers at scale and become instrumental in significantly advancing our understanding of connectomes in both the peripheral and the central nervous systems.

Computerized classification of intraductal breast lesions using histopathological images.

In the diagnosis of preinvasive breast cancer, some of the intraductal proliferations pose a special challenge. The continuum of intraductal breast lesions includes the usual ductal hyperplasia (UDH), atypical ductal hyperplasia (ADH), and ductal carcinoma in situ (DCIS). The current standard of care is to perform percutaneous needle biopsies for diagnosis of palpable and image-detected breast abnormalities. UDH is considered benign and patients diagnosed UDH undergo routine follow-up, whereas ADH and DCIS are considered actionable and patients diagnosed with these two subtypes get additional surgical procedures. About 250,000 new cases of intraductal breast lesions are diagnosed every year. A conservative estimate would suggest that at least 50% of these patients are needlessly undergoing unnecessary surgeries. Thus, improvement in the diagnostic reproducibility and accuracy is critically important for effective clinical management of these patients. In this study, a prototype system for automatically classifying breast microscopic tissues to distinguish between UDH and actionable subtypes (ADH and DCIS) is introduced. This system automatically evaluates digitized slides of tissues for certain cytological criteria and classifies the tissues based on the quantitative features derived from the images. The system is trained using a total of 327 regions of interest (ROIs) collected across 62 patient cases and tested with a sequestered set of 149 ROIs collected across 33 patient cases. An overall accuracy of 87.9% is achieved on the entire test data. The test accuracy of 84.6% is obtained with borderline cases (26 of the 33 test cases) only, when compared against the diagnostic accuracies of nine pathologists on the same set (81.2% average), indicates that the system is highly competitive with the expert pathologists as a stand-alone diagnostic tool and has a great potential in improving diagnostic accuracy and reproducibility when used as a "second reader" in conjunction with the pathologists.

Discovering the unknown: detection of emerging pathogens using a label-free light-scattering system.

A recently introduced technique for pathogen recognition called BARDOT (BActeria Rapid Detection using Optical scattering Technology) belongs to the broad class of optical sensors and relies on forward-scatter phenotyping (FSP). The specificity of FSP derives from the morphological information that bacterial material encodes on a coherent optical wavefront passing through the colony. The system collects elastically scattered light patterns that, given a constant environment, are unique to each bacterial species and serovar. The notable similarity between FSP technology and spectroscopies is their reliance on statistical machine learning to perform recognition. Currently used methods utilize traditional supervised techniques which assume completeness of training libraries. However, this restrictive assumption is known to be false for most experimental conditions, resulting in unsatisfactory levels of accuracy, poor specificity, and consequently limited overall performance for biodetection and classification tasks. The presented work demonstrates application of the BARDOT system to classify bacteria belonging to the Salmonella class in a nonexhaustive framework, that is, without full knowledge about all the possible classes that can be encountered. Our study uses a Bayesian approach to learning with a nonexhaustive training dataset to allow for the automated detection of unknown bacterial classes.

Multiple-instance learning algorithms for computer-aided detection.

Many computer-aided diagnosis (CAD) problems can be best modelled as a multiple-instance learning (MIL) problem with unbalanced data, i.e., the training data typically consists of a few positive bags, and a very large number of negative instances. Existing MIL algorithms are much too computationally expensive for these datasets. We describe CH, a framework for learning a Convex Hull representation of multiple instances that is significantly faster than existing MIL algorithms. Our CH framework applies to any standard hyperplane-based learning algorithm, and for some algorithms, is guaranteed to find the global optimal solution. Experimental studies on two different CAD applications further demonstrate that the proposed algorithm significantly improves diagnostic accuracy when compared to both MIL and traditional classifiers. Although not designed for standard MIL problems (which have both positive and negative bags and relatively balanced datasets), comparisons against other MIL methods on benchmark problems also indicate that the proposed method is competitive with the state-of-the-art.