Deep Learning News and Research

RSS

PromptSE AI predicts drug side effects by reasoning through biological clues

PromptSE uses structured LLM prompting to generate pharmacologically relevant side-effect representations, then feeds them into deep learning models to predict drug-side effect associations. In benchmark testing, PromptSE improved prediction performance, while PromptSE+ further boosted results by adding multi-modal drug information and graph-based refinement for sparse data.

1 Jun 2026

Can AI help build a fairer food system from farm to plate?

AI could help make food systems more personalized and climate-resilient by linking nutrition, health, farming, logistics, and sustainability data. The review highlights promise in precision nutrition, vertical farming, traceability, and smarter supply chains, while warning that privacy, equity, energy use, data silos, and governance gaps must be addressed.

27 May 2026

How the right diet could support the gut bacteria that matter most

This review examines how gut keystone bacteria shape microbiome stability, disease risk, and responses to diet. It highlights why better identification methods, experimental validation, and precision diet-probiotic strategies are needed to restore gut homeostasis.

26 May 2026

Deep learning model predicts vascular cognitive impairment from brain scans

The team built a DenseNet – a densely connected convolutional neural network – that learns hierarchical features directly from raw DTI volumes.

26 May 2026

New AI model detects hidden antibiotic resistance genes beyond standard databases

resLens uses genomic language models to detect antibiotic resistance genes from DNA sequences, performing especially well on long-read data and on some ARGs poorly represented in reference databases.

25 May 2026

Explainable AI could make breast cancer drug predictions safer and clearer

AI-driven genomic analysis could help researchers identify existing drugs that may be repurposed for subtype-specific breast cancer treatment. The review proposes an interpretability-driven framework that links multi-omics data, mechanistic validation, and clinical translation to make AI predictions more transparent and clinically useful.

22 May 2026

How an AI system learned to write expert-level scientific code

ERA is an AI system that uses large language models and tree search to automatically write, test, and refine scientific software for measurable research tasks. Across bioinformatics, COVID-19 forecasting, time-series analysis, geospatial modeling, neuroscience, and numerical computation, ERA produced expert-level solutions that often surpassed established human-developed methods.

20 May 2026

Advanced AI architectures improve meningioma detection on MRI scans

Meningiomas are the most common brain cancers arising from the protective soft tissue cover of the brain called the meninges.

19 May 2026

AI-powered handheld microscope may improve early cancer detection

Researchers at Rice University and The University of Texas MD Anderson Cancer Center have developed a compact, artificial intelligence-powered imaging device that could transform how clinicians detect cancer.

12 May 2026

Interactive AI model improves explainable lung cancer diagnosis from CT scans

Lung cancer remains one of the leading causes of cancer-related deaths worldwide, making early and accurate diagnosis essential for improving patient outcomes.

12 May 2026

Using AI to improve success rates of fecal microbiota transplants

MOZAIC leverages deep learning to optimize fecal microbiota transplantation, enhancing treatment efficacy through precise donor-recipient microbiome matching.

12 May 2026

New AI framework predicts how cells make fate decisions

Scientists at the Stowers Institute for Medical Research, Helmholtz Munich, the Technical University of Munich, and the University of Oxford have developed RegVelo - a new AI framework that simultaneously models cellular dynamics and gene regulation, enabling researchers to predict, simulate, and experimentally validate how cells make fate decisions.

11 May 2026

Qureight establishes Scientific Advisory Board focused on pulmonary hypertension

Qureight, an end-to-end imaging CRO that provides enterprise-grade imaging and precision endpoints for clinical trials with a focus on lung and heart disease, today announced it has established a Scientific Advisory Board (SAB) focused on pulmonary hypertension (PH).

From Qureight 11 May 2026

Suppressing TGF-β signaling reactivates dormant bone cells to treat osteoporosis

Osteoporosis, a common skeletal disorder, is marked by reduced bone mass and deterioration of bone microarchitecture, making bones more porous and prone to fractures.

8 May 2026

AI-powered MRI scans predict diabetes and heart disease risk

Research shows AI-enhanced MRI body composition metrics can predict diabetes and heart disease risks, offering insights for better health screening practices.

8 May 2026

Blocking TGF-beta may enhance osteoporosis bone regeneration therapies

A recent mouse model based study suggests that blocking TGF-β may improve osteoporosis treatment by helping quiescent osteoblasts in inactive bone surfaces return to an active state.

7 May 2026

Muscle fat levels on MRI linked to cardiometabolic risk

Using a deep learning model to analyze the composition of large muscles on MRI, German researchers found that the proportions of intermuscular fat and lean muscle mass were associated with high blood pressure and unhealthy lipid and blood sugar levels.

5 May 2026

AI creates detailed map of fat and muscle distribution to predict major health risks

Researchers used AI to analyze whole-body MRI scans from more than 66,000 participants to create the most detailed reference map to date of how fat and muscle are distributed in the human body across age, sex and height.

5 May 2026

AI model reads genetic code to trace evolutionary history

Researchers at the University of Oregon have developed an artificial intelligence tool that can read genetic code the way large language models like ChatGPT read text. Scanning the genome for biological mutation patterns, the computer model traces pairs of genes back in time to their last common ancestor.

4 May 2026

New tools for quicker and easier Mpox diagnosis

Following the rise in monkeypox (Mpox) cases, particularly in countries where the disease had not traditionally been observed, the World Health Organization (WHO) declared a Public Health Emergency of International Concern in June 2022 and again in August 2024.

4 May 2026