Improving image classification of gastrointestinal endoscopy using curriculum self-supervised learning.

Endoscopy, a widely used medical procedure for examining the gastrointestinal (GI) tract to detect potential disorders, poses challenges in manual diagnosis due to non-specific symptoms and difficulties in accessing affected areas. While supervised machine learning models have proven effective in assisting clinical diagnosis of GI disorders, the scarcity of image-label pairs created by medical experts limits their availability. To address these limitations, we propose a curriculum self-supervised learning framework inspired by human curriculum learning. Our approach leverages the HyperKvasir dataset, which comprises 100k unlabeled GI images for pre-training and 10k labeled GI images for fine-tuning. By adopting our proposed method, we achieved an impressive top-1 accuracy of 88.92% and an F1 score of 73.39%. This represents a 2.1% increase over vanilla SimSiam for the top-1 accuracy and a 1.9% increase for the F1 score. The combination of self-supervised learning and a curriculum-based approach demonstrates the efficacy of our framework in advancing the diagnosis of GI disorders. Our study highlights the potential of curriculum self-supervised learning in utilizing unlabeled GI tract images to improve the diagnosis of GI disorders, paving the way for more accurate and efficient diagnosis in GI endoscopy.

Improving long COVID-related text classification: a novel end-to-end domain-adaptive paraphrasing framework.

The emergence of long COVID during the ongoing COVID-19 pandemic has presented considerable challenges for healthcare professionals and researchers. The task of identifying relevant literature is particularly daunting due to the rapidly evolving scientific landscape, inconsistent definitions, and a lack of standardized nomenclature. This paper proposes a novel solution to this challenge by employing machine learning techniques to classify long COVID literature. However, the scarcity of annotated data for machine learning poses a significant obstacle. To overcome this, we introduce a strategy called medical paraphrasing, which diversifies the training data while maintaining the original content. Additionally, we propose a Data-Reweighting-Based Multi-Level Optimization Framework for Domain Adaptive Paraphrasing, supported by a Meta-Weight-Network (MWN). This innovative approach incorporates feedback from the downstream text classification model to influence the training of the paraphrasing model. During the training process, the framework assigns higher weights to the training examples that contribute more effectively to the downstream task of long COVID text classification. Our findings demonstrate that this method substantially improves the accuracy and efficiency of long COVID literature classification, offering a valuable tool for physicians and researchers navigating this complex and ever-evolving field.

Brain tumor classification based on neural architecture search.

Brain tumor is a life-threatening disease and causes about 0.25 million deaths worldwide in 2020. Magnetic Resonance Imaging (MRI) is frequently used for diagnosing brain tumors. In medically underdeveloped regions, physicians who can accurately diagnose and assess the severity of brain tumors from MRI are highly lacking. Deep learning methods have been developed to assist physicians in detecting brain tumors from MRI and determining their subtypes. In existing methods, neural architectures are manually designed by human experts, which is time-consuming and labor-intensive. To address this problem, we propose to automatically search for high-performance neural architectures for classifying brain tumors from MRIs, by leveraging a Learning-by-Self-Explanation (LeaSE) architecture search method. LeaSE consists of an explainer model and an audience model. The explainer aims at searching for a highly performant architecture by encouraging the architecture to generate high-fidelity explanations of prediction outcomes, where explanations' fidelity is evaluated by the audience model. LeaSE is formulated as a four-level optimization problem involving a sequence of four learning stages which are conducted end-to-end. We apply LeaSE for MRI-based brain tumor classification, including four classes: glioma, meningioma, pituitary tumor, and healthy, on a dataset containing 3264 MRI images. Results show that our method can search for neural architectures that achieve better classification accuracy than manually designed deep neural networks while having fewer model parameters. For example, our method achieves a test accuracy of 90.6% and an AUC of 95.6% with 3.75M parameters while the accuracy and AUC of a human-designed network-ResNet101-is 84.5% and 90.1% respectively with 42.56M parameters. In addition, our method outperforms state-of-the-art neural architecture search methods.

Pathway-Activity Likelihood Analysis and Metabolite Annotation for Untargeted Metabolomics Using Probabilistic Modeling.

Motivation: Untargeted metabolomics comprehensively characterizes small molecules and elucidates activities of biochemical pathways within a biological sample. Despite computational advances, interpreting collected measurements and determining their biological role remains a challenge. Results: To interpret measurements, we present an inference-based approach, termed Probabilistic modeling for Untargeted Metabolomics Analysis (PUMA). Our approach captures metabolomics measurements and the biological network for the biological sample under study in a generative model and uses stochastic sampling to compute posterior probability distributions. PUMA predicts the likelihood of pathways being active, and then derives probabilistic annotations, which assign chemical identities to measurements. Unlike prior pathway analysis tools that analyze differentially active pathways, PUMA defines a pathway as active if the likelihood that the path generated the observed measurements is above a particular (user-defined) threshold. Due to the lack of "ground truth" metabolomics datasets, where all measurements are annotated and pathway activities are known, PUMA is validated on synthetic datasets that are designed to mimic cellular processes. PUMA, on average, outperforms pathway enrichment analysis by 8%. PUMA is applied to two case studies. PUMA suggests many biological meaningful pathways as active. Annotation results were in agreement to those obtained using other tools that utilize additional information in the form of spectral signatures. Importantly, PUMA annotates many measurements, suggesting 23 chemical identities for metabolites that were previously only identified as isomers, and a significant number of additional putative annotations over spectral database lookups. For an experimentally validated 50-compound dataset, annotations using PUMA yielded 0.833 precision and 0.676 recall.

A Framework for Sensorimotor Cross-Perception and Cross-Behavior Knowledge Transfer for Object Categorization.

From an early age, humans learn to develop an intuition for the physical nature of the objects around them by using exploratory behaviors. Such exploration provides observations of how objects feel, sound, look, and move as a result of actions applied on them. Previous works in robotics have shown that robots can also use such behaviors (e.g., lifting, pressing, shaking) to infer object properties that camera input alone cannot detect. Such learned representations are specific to each individual robot and cannot currently be transferred directly to another robot with different sensors and actions. Moreover, sensor failure can cause a robot to lose a specific sensory modality which may prevent it from using perceptual models that require it as input. To address these limitations, we propose a framework for knowledge transfer across behaviors and sensory modalities such that: (1) knowledge can be transferred from one or more robots to another, and, (2) knowledge can be transferred from one or more sensory modalities to another. We propose two different models for transfer based on variational auto-encoders and encoder-decoder networks. The main hypothesis behind our approach is that if two or more robots share multi-sensory object observations of a shared set of objects, then those observations can be used to establish mappings between multiple features spaces, each corresponding to a combination of an exploratory behavior and a sensory modality. We evaluate our approach on a category recognition task using a dataset in which a robot used 9 behaviors, coupled with 4 sensory modalities, performed multiple times on 100 objects. The results indicate that sensorimotor knowledge about objects can be transferred both across behaviors and across sensory modalities, such that a new robot (or the same robot, but with a different set of sensors) can bootstrap its category recognition models without having to exhaustively explore the full set of objects.