Medical AI brings together our reporting on machine learning for clinical and biomedical problems, from diagnosis and prognosis to medical image analysis and decision support. Because mistakes in this domain carry a human cost, we pay close attention to evaluation, calibration, uncertainty, and the gap between benchmark numbers and bedside reliability. Expect grounded explainers of recent research rather than uncritical product announcements.
Picture an older woman who lives alone and is frail but fiercely independent. Her nearest relative is hours away. Her doctor would love to know whether she still climbs her stairs each day, whether she sits more than she did…
Wearable Activity Recognition Models Compared Read More »
A patient sits under a dermatoscope while a clinician tries to decide, in under a minute, whether an irregular brown patch on their shoulder is an ordinary mole or something that needs a biopsy. That judgment call happens millions of…
Quantum Layer Boosts CNN Melanoma Classification Accuracy Read More »
A radiologist in a district hospital reads sixty chest radiographs before lunch. Somewhere in that stack is a faint patch of haze near an upper lobe, the kind of shadow that could be old scarring, a passing infection, or early…
ETDHDNet Tuberculosis Detection From Chest Radiographs Read More »
A nurse on a medical caravan outside Fez presses a handheld retinal camera against a patient’s eye, taps a phone screen and waits. About a second later the app answers that diabetic retinopathy is present, with 90 percent confidence. The…
Diabetic Retinopathy Detection on a Phone, Audited Read More »
Inside an oncology department preparing a patient for image-guided adaptive radiation therapy, or inside a musculoskeletal clinic evaluating a torn meniscus, lying motionless inside a magnetic resonance bore for thirty minutes is an exhausting ordeal. Acquiring crisp, high-resolution T2-weighted brain…
High Frequency Modulated Transformer for MRI Super-Resolution Read More »
Inside a hospital neuroimaging suite, a patient recovering from a traumatic brain injury who has developed post-traumatic epilepsy (PTE) lies inside a 3 Tesla magnetic resonance scanner equipped with a 20-channel head coil[cite: 49]. Standard structural T1-weighted scans can reveal…
Subspace and Self Supervised Denoising for MRSI Brain Scans Read More »
Skin cancer detection has become one of the clearer success stories for deep learning in medicine, with models trained on large dermoscopy image collections repeatedly matching or approaching dermatologist level accuracy on benchmark datasets. The catch is that the models…
SSD-KD: A Compact Skin Lesion Classifier That Outperforms Its Own Teacher Model Read More »
Inside a microfluidic chamber sealed atop a 90-nanometer CMOS biochip, an elongated nanoscale particle—a 300-nanometer nanoplastic fiber, a rod-shaped bacteriophage, or a double-stranded nucleic acid bundle—drifts in a 100-millimolar saline solution just 40 nanometers above a grid of gold-copper nanoelectrodes.…
ML and Bayesian Metrology for Nanoscale Capacitance Sensors Read More »
Predicting how a cancer patient’s disease will progress from a combination of pathology images and molecular data is a well established goal in computational pathology. The typical approach processes the whole slide image through one pipeline, processes the gene expression…
Graph Attention Fusion of Pathology Images and Gene Expression Predicts Cancer Survival Read More »
The retina is one of the only places in the human body where blood vessels can be photographed directly without surgery, which is exactly why so much diagnostic attention focuses on it. Vessels enter and exit through the optic nerve…
How MaskVSC Reconnects Broken Retinal Blood Vessels Read More »