Electrocardiogram Instruction Tuning for Report Generation

Electrocardiogram (ECG) serves as the primary non-invasive diagnostic tool for cardiac conditions monitoring, are crucial in assisting clinicians. Recent studies have concentrated on classifying cardiac conditions using ECG data but have overlooked ECG report generation, which is not only time-consuming but also requires clinical expertise. To automate ECG report generation and ensure its versatility, we propose the Multimodal ECG Instruction Tuning (MEIT) framework, the \textit{first} attempt to tackle ECG report generation with LLMs and multimodal instructions. To facilitate future research, we establish a benchmark to evaluate MEIT with various LLMs backbones across two large-scale ECG datasets. Our approach uniquely aligns the representations of the ECG signal and the report, and we conduct extensive experiments to benchmark MEIT with nine open source LLMs, using more than 800,000 ECG reports. MEIT's results underscore the superior performance of instruction-tuned LLMs, showcasing their proficiency in quality report generation, zero-shot capabilities, and resilience to signal perturbation. These findings emphasize the efficacy of our MEIT framework and its potential for real-world clinical application.

Multi-View Variational Autoencoder for Missing Value Imputation in Untargeted Metabolomics

Background: Missing data is a common challenge in mass spectrometry-based metabolomics, which can lead to biased and incomplete analyses. The integration of whole-genome sequencing (WGS) data with metabolomics data has emerged as a promising approach to enhance the accuracy of data imputation in metabolomics studies. Method: In this study, we propose a novel method that leverages the information from WGS data and reference metabolites to impute unknown metabolites. Our approach utilizes a multi-view variational autoencoder to jointly model the burden score, polygenetic risk score (PGS), and linkage disequilibrium (LD) pruned single nucleotide polymorphisms (SNPs) for feature extraction and missing metabolomics data imputation. By learning the latent representations of both omics data, our method can effectively impute missing metabolomics values based on genomic information. Results: We evaluate the performance of our method on empirical metabolomics datasets with missing values and demonstrate its superiority compared to conventional imputation techniques. Using 35 template metabolites derived burden scores, PGS and LD-pruned SNPs, the proposed methods achieved r2-scores > 0.01 for 71.55% of metabolites. Conclusion: The integration of WGS data in metabolomics imputation not only improves data completeness but also enhances downstream analyses, paving the way for more comprehensive and accurate investigations of metabolic pathways and disease associations. Our findings offer valuable insights into the potential benefits of utilizing WGS data for metabolomics data imputation and underscore the importance of leveraging multi-modal data integration in precision medicine research.

FedAIoT: A Federated Learning Benchmark for Artificial Intelligence of Things

There is a significant relevance of federated learning (FL) in the realm of Artificial Intelligence of Things (AIoT). However, most existing FL works are not conducted on datasets collected from authentic IoT devices that capture unique modalities and inherent challenges of IoT data. In this work, we introduce FedAIoT, an FL benchmark for AIoT to fill this critical gap. FedAIoT includes eight datatsets collected from a wide range of IoT devices. These datasets cover unique IoT modalities and target representative applications of AIoT. FedAIoT also includes a unified end-to-end FL framework for AIoT that simplifies benchmarking the performance of the datasets. Our benchmark results shed light on the opportunities and challenges of FL for AIoT. We hope FedAIoT could serve as an invaluable resource to foster advancements in the important field of FL for AIoT. The repository of FedAIoT is maintained at https://github.com/AIoT-MLSys-Lab/FedAIoT.

Consistency of Lloyd's Algorithm Under Perturbations

In the context of unsupervised learning, Lloyd's algorithm is one of the most widely used clustering algorithms. It has inspired a plethora of work investigating the correctness of the algorithm under various settings with ground truth clusters. In particular, in 2016, Lu and Zhou have shown that the mis-clustering rate of Lloyd's algorithm on $n$ independent samples from a sub-Gaussian mixture is exponentially bounded after $O(\log(n))$ iterations, assuming proper initialization of the algorithm. However, in many applications, the true samples are unobserved and need to be learned from the data via pre-processing pipelines such as spectral methods on appropriate data matrices. We show that the mis-clustering rate of Lloyd's algorithm on perturbed samples from a sub-Gaussian mixture is also exponentially bounded after $O(\log(n))$ iterations under the assumptions of proper initialization and that the perturbation is small relative to the sub-Gaussian noise. In canonical settings with ground truth clusters, we derive bounds for algorithms such as $k$-means$++$ to find good initializations and thus leading to the correctness of clustering via the main result. We show the implications of the results for pipelines measuring the statistical significance of derived clusters from data such as SigClust. We use these general results to derive implications in providing theoretical guarantees on the misclustering rate for Lloyd's algorithm in a host of applications, including high-dimensional time series, multi-dimensional scaling, and community detection for sparse networks via spectral clustering.

Evaluating the Robustness to Instructions of Large Language Models

Recently, Instruction fine-tuning has risen to prominence as a potential method for enhancing the zero-shot capabilities of Large Language Models (LLMs) on novel tasks. This technique has shown an exceptional ability to boost the performance of moderately sized LLMs, sometimes even reaching performance levels comparable to those of much larger model variants. The focus is on the robustness of instruction-tuned LLMs to seen and unseen tasks. We conducted an exploration of six models including Alpaca, Vicuna, WizardLM, and Traditional Task-oriented Models(Flan-T5-XL/XXL, T0++) using real-world relation extraction datasets as case studies. We carried out a comprehensive evaluation of these instruction-following LLMs which have been tuned based on open-domain instructions and task-oriented instructions. The main discussion is their performance and robustness towards instructions. We have observed that in most cases, the model's performance in dealing with unfamiliar instructions tends to worsen significantly, and the robustness of the model for RE instructions deteriorates compared to QA. Further, we discovered that up until a certain parameter size threshold (3B), the performance of the FLAN-T5 model improves as the parameter count increases. The robustness of different scales of FLAN-T5 models to RE instruction is worse than the robustness to QA instruction.

CLCLSA: Cross-omics Linked embedding with Contrastive Learning and Self Attention for multi-omics integration with incomplete multi-omics data

Integration of heterogeneous and high-dimensional multi-omics data is becoming increasingly important in understanding genetic data. Each omics technique only provides a limited view of the underlying biological process and integrating heterogeneous omics layers simultaneously would lead to a more comprehensive and detailed understanding of diseases and phenotypes. However, one obstacle faced when performing multi-omics data integration is the existence of unpaired multi-omics data due to instrument sensitivity and cost. Studies may fail if certain aspects of the subjects are missing or incomplete. In this paper, we propose a deep learning method for multi-omics integration with incomplete data by Cross-omics Linked unified embedding with Contrastive Learning and Self Attention (CLCLSA). Utilizing complete multi-omics data as supervision, the model employs cross-omics autoencoders to learn the feature representation across different types of biological data. The multi-omics contrastive learning, which is used to maximize the mutual information between different types of omics, is employed before latent feature concatenation. In addition, the feature-level self-attention and omics-level self-attention are employed to dynamically identify the most informative features for multi-omics data integration. Extensive experiments were conducted on four public multi-omics datasets. The experimental results indicated that the proposed CLCLSA outperformed the state-of-the-art approaches for multi-omics data classification using incomplete multi-omics data.

ImageNomer: developing an fMRI and omics visualization tool to detect racial bias in functional connectivity

It can be difficult to identify trends and perform quality control in large, high-dimensional fMRI or omics datasets. To remedy this, we develop ImageNomer, a data visualization and analysis tool that allows inspection of both subject-level and cohort-level features. The tool allows visualization of phenotype correlation with functional connectivity (FC), partial connectivity (PC), dictionary components (PCA and our own method), and genomic data (single-nucleotide polymorphisms, SNPs). In addition, it allows visualization of weights from arbitrary ML models. ImageNomer is built with a Python backend and a Vue frontend. We validate ImageNomer using the Philadelphia Neurodevelopmental Cohort (PNC) dataset, which contains multitask fMRI and SNP data of healthy adolescents. Using correlation, greedy selection, or model weights, we find that a set of 10 FC features can explain 15% of variation in age, compared to 35% for the full 34,716 feature model. The four most significant FCs are either between bilateral default mode network (DMN) regions or spatially proximal subcortical areas. Additionally, we show that whereas both FC (fMRI) and SNPs (genomic) features can account for 10-15% of intelligence variation, this predictive ability disappears when controlling for race. We find that FC features can be used to predict race with 85% accuracy, compared to 78% accuracy for sex prediction. Using ImageNomer, this work casts doubt on the possibility of finding unbiased intelligence-related features in fMRI and SNPs of healthy adolescents.

Multi-view information fusion using multi-view variational autoencoders to predict proximal femoral strength

Background and aim: Hip fracture can be devastating. The proximal femoral strength can be computed by subject-specific finite element (FE) analysis (FEA) using quantitative CT images. The aim of this paper is to design a deep learning-based model for hip fracture prediction with multi-view information fusion. Method: We developed a multi-view variational autoencoder (MMVAE) for feature representation learning and designed the product of expert model (PoE) for multi-view information fusion.We performed genome-wide association studies (GWAS) to select the most relevant genetic features with proximal femoral strengths and integrated genetic features with DXA-derived imaging features and clinical variables for proximal femoral strength prediction. Results: The designed model achieved the mean absolute percentage error of 0.2050,0.0739 and 0.0852 for linear fall, nonlinear fall and nonlinear stance fracture load prediction, respectively. For linear fall and nonlinear stance fracture load prediction, integrating genetic and DXA-derived imaging features were beneficial; while for nonlinear fall fracture load prediction, integrating genetic features, DXA-derived imaging features as well as clinical variables, the model achieved the best performance. Conclusion: The proposed model is capable of predicting proximal femoral strengths using genetic features, DXA-derived imaging features as well as clinical variables. Compared to performing FEA using QCT images to calculate proximal femoral strengths, the presented method is time-efficient and cost effective, and radiation dosage is limited. From the technique perspective, the final models can be applied to other multi-view information integration tasks.

A robust kernel machine regression towards biomarker selection in multi-omics datasets of osteoporosis for drug discovery

Many statistical machine approaches could ultimately highlight novel features of the etiology of complex diseases by analyzing multi-omics data. However, they are sensitive to some deviations in distribution when the observed samples are potentially contaminated with adversarial corrupted outliers (e.g., a fictional data distribution). Likewise, statistical advances lag in supporting comprehensive data-driven analyses of complex multi-omics data integration. We propose a novel non-linear M-estimator-based approach, "robust kernel machine regression (RobKMR)," to improve the robustness of statistical machine regression and the diversity of fictional data to examine the higher-order composite effect of multi-omics datasets. We address a robust kernel-centered Gram matrix to estimate the model parameters accurately. We also propose a robust score test to assess the marginal and joint Hadamard product of features from multi-omics data. We apply our proposed approach to a multi-omics dataset of osteoporosis (OP) from Caucasian females. Experiments demonstrate that the proposed approach effectively identifies the inter-related risk factors of OP. With solid evidence (p-value = 0.00001), biological validations, network-based analysis, causal inference, and drug repurposing, the selected three triplets ((DKK1, SMTN, DRGX), (MTND5, FASTKD2, CSMD3), (MTND5, COG3, CSMD3)) are significant biomarkers and directly relate to BMD. Overall, the top three selected genes (DKK1, MTND5, FASTKD2) and one gene (SIDT1 at p-value= 0.001) significantly bond with four drugs- Tacrolimus, Ibandronate, Alendronate, and Bazedoxifene out of 30 candidates for drug repurposing in OP. Further, the proposed approach can be applied to any disease model where multi-omics datasets are available.