A cross-disciplinary team from Sichuan University's NICUs employs a machine learning pipeline to classify neonatal intestinal diseases using bowel sound recordings captured by a digital stethoscope. They preprocess audio with filters, extract time–frequency features such as MFCCs, and train a transformer-based model combined with a Random Forest to detect conditions like NEC, FPIAP, and obstruction, aiming to supplement subjective clinical assessment with objective, automated diagnostics.

Key points

• Collected neonatal bowel sounds via 3M Littmann 3200 digital stethoscope with 2-minute recordings from six abdominal regions, filtered to exclude noise exceeding 30%.
• Extracted acoustic features—zero-crossing rate, spectral centroid, chroma, MFCCs—after pre-emphasis, framing, and Hamming windowing, forming a multidimensional feature vector.
• Trained a Random Forest for disease detection and a transformer-based network for multi-class classification (NEC, FPIAP, volvulus, obstruction), validated via tenfold cross-validation and external cohorts with high AUC.

Why it matters: An AI-based bowel sound diagnostic tool offers rapid, noninvasive neonatal intestinal disease screening, potentially reducing delays and improving outcomes compared with subjective auscultation.

The collaboration between biotech company Klothea Bio and Healthy Longevity Clinic leverages a proprietary mRNA platform to upregulate endogenous Klotho, an anti-aging protein. By delivering lipid-encapsulated mRNA instructions, they activate the body’s protein synthesis pathways. The approach aims to counteract age-associated inflammation, cognitive decline, and immune deterioration in adults.

Key points

• mRNA-based Klotho delivery via lipid nanoparticles upregulates anti-aging protein expression.
• Collaboration recognized as Top 40 semifinalist in the $101 million XPRIZE Healthspan milestone competition.
• Preclinical data suggest improved cognition, reduced inflammation, and enhanced immune function in aging models.

Why it matters: Elevating Klotho via mRNA could revolutionize aging therapeutics by restoring cellular homeostasis and delaying multiple age-related disorders.

Researchers at Taizhou Cancer Hospital leverage MRI-based radiomics and machine learning to classify high-grade glioma grades and forecast overall survival. They extract 107 quantitative features from T1-weighted images, perform LASSO feature selection, balance data with SMOTE, and compare classifiers—finding that XGBoost and a stacking fusion model yield top performance metrics.

Key points

• Extracted 107 MRI radiomics features (first-order, shape, texture) and filtered for ICC>0.90 repeatability.
• Applied LASSO for dimensionality reduction, SMOTE to balance classes, and compared six classifiers; XGBoost achieved top non-fusion performance.
• Developed a stacking fusion ensemble yielding AUC=0.95, with SHAP highlighting texture metrics (SizeZoneNonUniformity, InverseVariance) as key prognostic indicators.

Why it matters: This study demonstrates a robust AI radiomics framework that noninvasively grades gliomas and forecasts survival, advancing personalized oncology and reducing reliance on risky biopsies.

Firat University’s digital forensics and neuroscience researchers introduce FriendPat, a new-generation explainable feature engineering model for EEG-based epilepsy detection. FriendPat computes channel distance matrices, applies voting-based feature extraction, and employs CWINCA feature selection with a t-algorithm kNN classifier. Integrated with Directed Lobish symbolic language, it produces interpretable connectomes for accurate epilepsy diagnosis.

Key points

• FriendPat uses L1-norm channel distance matrices and pivot-based voting to generate 595-dimensional feature vectors from 35-channel EEG signals.
• CWINCA self-organized selector reduces features to 82 through cumulative weight thresholds, ensuring linear time complexity and optimal feature subset.
• tkNN ensemble classifier coupled with Directed Lobish symbolism achieves 99.61% accuracy under 10-fold CV and generates interpretable cortical connectome diagrams.

Why it matters: This explainable, lightweight EEG classification approach could transform clinical epilepsy diagnostics by combining high accuracy with interpretable neural connectome insights.

Longevity Method, a pioneering wellness firm, debuts five research-backed anti-aging formulations. Each targets core aging pathways—NAD+ restoration, sirtuin activation, mitochondrial support, sleep quality, muscle endurance, and hormonal balance—to extend healthspan and vitality.

Key points

• Live Longer NMN+ replenishes NAD+ with NMN, Urolithin A, Ca-AKG, TMG to boost mitochondrial energy.
• Live Longer Booster blends Resveratrol, Quercetin, Fisetin, Spermidine to activate sirtuins and clear senescent cells.
• Sleep Better, Play Harder, and Keep Balanced use targeted botanicals and micronutrients for restorative sleep, muscle endurance, and hormonal balance.

Axtria Inc., renowned for life sciences data analytics, partners with Genloop to deploy domain-trained agentic AI models. These LLM-based agents leverage institutional knowledge, integrate seamlessly into enterprise infrastructures, and navigate regulatory requirements to enhance precision and workflow efficiency in pharmaceutical applications.

Key points

• Domain-trained LLMs fine-tuned on life sciences workflows deliver contextualized output accuracy.
• Agentic AI agents integrate via secure APIs into CRM and ERP systems for seamless deployment.
• Platform embeds regulatory compliance checks and audit logs to meet FDA and EMA requirements.

Why it matters: Domain-trained agentic AI reduces development costs and regulatory risks while enhancing data-driven decision making across life sciences.

Independent reporter Holly Baxter profiles women who adopt advanced biohacking tools—from PEMF and intranasal red-light devices to personalised nutrition and monitoring—to optimise bone strength, sleep, and metabolic health and prepare for menopause.

Key points

• Women integrate PEMF mats at home for inflammation reduction and cellular recovery, inspired by veterinary uses.
• Advanced biohacking tech like NanoVi oxygen delivery, intranasal photobiomodulation, and LYMA lasers support brain, skin, and hair protocols.
• Biomarker tracking via Oura rings and bioelectrical impedance scales guides personalized nutrition, exercise, and hormone-cycle-based therapies.