In an increasingly data-driven world, artificial intelligence is expected to be a key tool for converting big data into tangible benefits and the healthcare domain is no exception to this. Machine learning aims to identify complex patterns in multi-dimensional data and use these uncovered patterns to classify new unseen cases or make data-driven predictions. In recent years, deep neural networks have shown to be capable of producing results that considerably exceed those of conventional machine learning methods for various classification and regression tasks. In this paper, we provide an accessible tutorial of the most important supervised machine learning concepts and methods, including deep learning, which are potentially the most relevant for the medical domain. We aim to take some of the mystery out of machine learning and depict how machine learning models can be useful for medical applications. Finally, this tutorial provides a few practical suggestions for how to properly design a machine learning model for a generic medical problem.

Non-contrast computed tomography (NCCT) is commonly used for volumetric follow-up assessment of ischemic strokes. However, manual lesion segmentation is time-consuming and subject to high inter-observer variability. The aim of this study was to develop and establish a baseline convolutional neural network (CNN) model for automatic NCCT lesion segmentation. A total of 252 multi-center clinical NCCT datasets, acquired from 22 centers, and corresponding manual segmentations were used to train (204 datasets) and validate (48 datasets) a 3D multi-scale CNN model for lesion segmentation. Post-processing methods were implemented to improve the CNN-based lesion segmentations. The final CNN model and post-processing method was evaluated using 39 out-of-distribution holdout test datasets, acquired at seven centers that did not contribute to the training or validation datasets. Each test image was segmented by two or three neuroradiologists. The Dice similarity coefficient (DSC) and predicted lesion volumes were used to evaluate the segmentations. The CNN model achieved a mean DSC score of 0.47 on the validation NCCT datasets. Post-processing significantly improved the DSC to 0.50 (P < 0.01). On the holdout test set, the CNN model achieved a mean DSC score of 0.42, which was also significantly improved to 0.45 (P < 0.05) by post-processing. Importantly, the automatically segmented lesion volumes were not significantly different from the lesion volumes determined by the expert observers (P > 0.05) and showed excellent agreement with manual lesion segmentation volumes (intraclass correlation coefficient, ICC = 0.88). The proposed CNN model can automatically and reliably segment ischemic stroke lesions in clinical NCCT datasets. Post-processing techniques can further improve accuracy. As the model was trained and evaluated on datasets from multiple centers, it is broadly applicable and is publicly available. INDEX TERMS Artificial neural networks, brain, computed tomography, computer-assisted image analysis, convolutional neural networks, deep learning, machine learning, stroke.

BackgroundLesion-symptom mapping (LSM) is a statistical technique to investigate the population-specific relationship between structural integrity and post-stroke clinical outcome. In clinical practice, patients are commonly evaluated using the National Institutes of Health Stroke Scale (NIHSS), an 11-domain clinical score to quantitate neurological deficits due to stroke. So far, LSM studies have mostly used the total NIHSS score for analysis, which might not uncover subtle structure–function relationships associated with the specific sub-domains of the NIHSS evaluation. Thus, the aim of this work was to investigate the feasibility to perform LSM analyses with sub-score information to reveal category-specific structure–function relationships that a total score may not reveal.MethodsEmploying a multivariate technique, LSM analyses were conducted using a sample of 180 patients with NIHSS assessment at 48-hour post-stroke from the ESCAPE trial. The NIHSS domains were grouped into six categories using two schemes. LSM was conducted for each category of the two groupings and the total NIHSS score.ResultsSub-score LSMs not only identify most of the brain regions that are identified as critical by the total NIHSS score but also reveal additional brain regions critical to each function category of the NIHSS assessment without requiring extensive, specialised assessments.ConclusionThese findings show that widely available sub-scores of clinical outcome assessments can be used to investigate more specific structure–function relationships, which may improve predictive modelling of stroke outcomes in the context of modern clinical stroke assessments and neuroimaging.Trial registration numberNCT01778335.