Meta-learning in healthcare: A survey

As a subset of machine learning, meta-learning, or learning to learn, aims at improving the model's capabilities by employing prior knowledge and experience. A meta-learning paradigm can appropriately tackle the conventional challenges of traditional learning approaches, such as insufficient number of samples, domain shifts, and generalization. These unique characteristics position meta-learning as a suitable choice for developing influential solutions in various healthcare contexts, where the available data is often insufficient, and the data collection methodologies are different. This survey discusses meta-learning broad applications in the healthcare domain to provide insight into how and where it can address critical healthcare challenges. We first describe the theoretical foundations and pivotal methods of meta-learning. We then divide the employed meta-learning approaches in the healthcare domain into two main categories of multi/single-task learning and many/few-shot learning and survey the studies. Finally, we highlight the current challenges in meta-learning research, discuss the potential solutions and provide future perspectives on meta-learning in healthcare.

RVD: A Handheld Device-Based Fundus Video Dataset for Retinal Vessel Segmentation

Retinal vessel segmentation is generally grounded in image-based datasets collected with bench-top devices. The static images naturally lose the dynamic characteristics of retina fluctuation, resulting in diminished dataset richness, and the usage of bench-top devices further restricts dataset scalability due to its limited accessibility. Considering these limitations, we introduce the first video-based retinal dataset by employing handheld devices for data acquisition. The dataset comprises 635 smartphone-based fundus videos collected from four different clinics, involving 415 patients from 50 to 75 years old. It delivers comprehensive and precise annotations of retinal structures in both spatial and temporal dimensions, aiming to advance the landscape of vasculature segmentation. Specifically, the dataset provides three levels of spatial annotations: binary vessel masks for overall retinal structure delineation, general vein-artery masks for distinguishing the vein and artery, and fine-grained vein-artery masks for further characterizing the granularities of each artery and vein. In addition, the dataset offers temporal annotations that capture the vessel pulsation characteristics, assisting in detecting ocular diseases that require fine-grained recognition of hemodynamic fluctuation. In application, our dataset exhibits a significant domain shift with respect to data captured by bench-top devices, thus posing great challenges to existing methods. In the experiments, we provide evaluation metrics and benchmark results on our dataset, reflecting both the potential and challenges it offers for vessel segmentation tasks. We hope this challenging dataset would significantly contribute to the development of eye disease diagnosis and early prevention.

Deep conv-attention model for diagnosing left bundle branch block from 12-lead electrocardiograms

Cardiac resynchronization therapy (CRT) is a treatment that is used to compensate for irregularities in the heartbeat. Studies have shown that this treatment is more effective in heart patients with left bundle branch block (LBBB) arrhythmia. Therefore, identifying this arrhythmia is an important initial step in determining whether or not to use CRT. On the other hand, traditional methods for detecting LBBB on electrocardiograms (ECG) are often associated with errors. Thus, there is a need for an accurate method to diagnose this arrhythmia from ECG data. Machine learning, as a new field of study, has helped to increase human systems' performance. Deep learning, as a newer subfield of machine learning, has more power to analyze data and increase systems accuracy. This study presents a deep learning model for the detection of LBBB arrhythmia from 12-lead ECG data. This model consists of 1D dilated convolutional layers. Attention mechanism has also been used to identify important input data features and classify inputs more accurately. The proposed model is trained and validated on a database containing 10344 12-lead ECG samples using the 10-fold cross-validation method. The final results obtained by the model on the 12-lead ECG data are as follows. Accuracy: 98.80+-0.08%, specificity: 99.33+-0.11 %, F1 score: 73.97+-1.8%, and area under the receiver operating characteristics curve (AUC): 0.875+-0.0192. These results indicate that the proposed model in this study can effectively diagnose LBBB with good efficiency and, if used in medical centers, will greatly help diagnose this arrhythmia and early treatment.

DeFungi: Direct Mycological Examination of Microscopic Fungi Images

Traditionally, diagnosis and treatment of fungal infections in humans depend heavily on face-to-face consultations or examinations made by specialized laboratory scientists known as mycologists. In many cases, such as the recent mucormycosis spread in the COVID-19 pandemic, an initial treatment can be safely suggested to the patient during the earliest stage of the mycological diagnostic process by performing a direct examination of biopsies or samples through a microscope. Computer-aided diagnosis systems using deep learning models have been trained and used for the late mycological diagnostic stages. However, there are no reference literature works made for the early stages. A mycological laboratory in Colombia donated the images used for the development of this research work. They were manually labelled into five classes and curated with a subject matter expert assistance. The images were later cropped and patched with automated code routines to produce the final dataset. This paper presents experimental results classifying five fungi types using two different deep learning approaches and three different convolutional neural network models, VGG16, Inception V3, and ResNet50. The first approach benchmarks the classification performance for the models trained from scratch, while the second approach benchmarks the classification performance using pre-trained models based on the ImageNet dataset. Using k-fold cross-validation testing on the 5-class dataset, the best performing model trained from scratch was Inception V3, reporting 73.2% accuracy. Also, the best performing model using transfer learning was VGG16 reporting 85.04%. The statistics provided by the two approaches create an initial point of reference to encourage future research works to improve classification performance. Furthermore, the dataset built is published in Kaggle and GitHub to foster future research.