This paper presents a novel machine learning framework for detecting Paroxysmal Atrial Fibrillation (PxAF), a pathological characteristic of Electrocardiogram (ECG) that can lead to fatal conditions such as heart attack. To enhance the learning process, the framework involves a Generative Adversarial Network (GAN) along with a Neural Architecture Search (NAS) in the data preparation and classifier optimization phases. The GAN is innovatively invoked to overcome the class imbalance of the training data by producing the synthetic ECG for PxAF class in a certified manner. The effect of the certified GAN is statistically validated. Instead of using a general-purpose classifier, the NAS automatically designs a highly accurate convolutional neural network architecture customized for the PxAF classification task. Experimental results show that the accuracy of the proposed framework exhibits a high value of 99% which not only enhances state-of-the-art by up to 5.1%, but also improves the classification performance of the two widely-accepted baseline methods, ResNet-18, and Auto-Sklearn, by 2.2% and 6.1%.

Evidence-based recommendations are often published in the form of clinical guidelines and protocols, as documents intended to be used by clinicians to provide the state of the art care. However, as demonstrated repeatedly in multiple clinical domains, clinicians often do not sufficiently adhere to the guidelines in a manner sensitive to the context of each patient. Such gaps are important to detect; fast, large-scale detection might lead to specific adjustments, usually of the clinicians' management patterns, but possibly of the guidelines themselves. In this study, we evaluated the DiscovErr system, in which we had implemented a new methodology for assessment of compliance to continuous implementation of clinical guidelines. This new methodology is based on a bi-directional search from the objective of the guideline to the longitudinal multivariate patient data, and vice versa. The evaluation of DiscovErr was performed in the type 2 Diabetes management domain, by comparing its performance to a panel of three clinicians, two experts in diabetes-patient management and a senior family practitioner highly experienced in diabetes treatment. The system and the three experts commented on the management of 10 patients who were randomly selected before the evaluation from a database containing longitudinal records of 2,000 type 2 diabetes patients. On average, each patient record spanned 5.23 years; the overall data of the selected patients included 1,584 time-oriented medical transactions (laboratory tests or medication administrations). We assessed the correctness (i.e.

Telementoring surgeons as they perform surgery can be essential in the treatment of patients when in situ expertise is not available. Nonetheless, expert mentors are often unavailable to provide trainees with real-time medical guidance. When mentors are unavailable, a fallback autonomous mechanism should provide medical practitioners with the required guidance. However, AI/autonomous mentoring in medicine has been limited by the availability of generalizable prediction models, and surgical procedures datasets to train those models with. This work presents the initial steps towards the development of an intelligent artificial system for autonomous medical mentoring. Specifically, we present the first Database for AI Surgical Instruction (DAISI). DAISI leverages on images and instructions to provide step-by-step demonstrations of how to perform procedures from various medical disciplines. The dataset was acquired from real surgical procedures and data from academic textbooks. We used DAISI to train an encoder-decoder neural network capable of predicting medical instructions given a current view of the surgery. Afterwards, the instructions predicted by the network were evaluated using cumulative BLEU scores and input from expert physicians. According to the BLEU scores, the predicted and ground truth instructions were as high as 67% similar. Additionally, expert physicians subjectively assessed the algorithm using Likert scale, and considered that the predicted descriptions were related to the images. This work provides a baseline for AI algorithms to assist in autonomous medical mentoring.

The Animal-AI Olympics, which will begin this June, aims to "benchmark the current level of various AIs against different animal species using a range of established animal cognition tasks." At stake are bragging rights and US $10,000 in prizes. The project, a partnership between the University of Cambridge's Leverhulme Centre for the Future of Intelligence and GoodAI, a research institution based in Prague, is a new way to evaluate the progress of AI systems toward what researchers call artificial general intelligence. Such an assessment is necessary, the organizers say, because recent benchmarks are somewhat deceiving. While AI systems have bested human grandmasters in a host of challenging competitions, including the board game Go and the video game StarCraft, these matchups only proved that the AIs were astoundingly good at those particular games.

An algorithm that uses artificial intelligence can automatically and more accurately diagnose a potentially devastating cause of childhood blindness than most expert physicians, a paper published in JAMA Ophthalmology suggests. The finding could help prevent blindness in more babies with the disease, called retinopathy of prematurity, or ROP. Musician Stevie Wonder went blind due to this condition. The algorithm accurately diagnosed the condition in images of infant eyes 91 percent of the time. On the other hand, a team of eight physicians with ROP expertise who examined the same images had an average accuracy rate of 82 percent.