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.
Teaching a model to outline a skin lesion or a polyp usually means paying experts to trace thousands of images by hand, which is slow and expensive. The obvious shortcut is to label a few and let the model learn…
SAUF-Net Separates Lesion Shape From Its Appearance Read More »
A segmentation model can score a near perfect Dice number and still be wrong in the one place a hematologist cares about. Draw the outline of a white blood cell a few pixels too wide and the overlap metric barely…
RABR-Net Refines Cell Boundaries That Dice Scores Miss Read More »
Some hearts have one working pump where the textbook says there should be two. Children born with single ventricle physiology grow up with a heart that mixes oxygen rich and oxygen poor blood, and a chamber that was supposed to…
SV-Cine Segments Single Ventricle Hearts From Cine MRI Read More »
A patient enrolled in a pancreatic cyst surveillance program gets an MRI this year and a CT the next, because that is simply how her care happened to unfold across two hospitals. A radiologist reviewing both wants consistent measurements of…
One Model Learns To Segment The Pancreas On CT And MRI Read More »
A haematologist bent over a microscope is doing something quietly heroic. From a smear of blood, they pick out the one cell in a thousand that looks wrong, the immature blast that can be the first sign of leukaemia, and…
CytoDiffusion: A Generative Classifier for Blood Cell Morphology Read More »
A tube of blood holds a faint rumour of what is happening deep inside the body. Among the debris floating in plasma are fragments of RNA shed by cells, and when a tumour is growing somewhere, it leaks its own…
Exai-1: A Cell Free RNA Foundation Model for Liquid Biopsy Read More »
This article covers an experimental medical device and animal study, not a clinical treatment. The system described here has been tested in laboratory setups and on porcine (pig) ex vivo tissue, not in living patients. Nothing in this article is…
A Magnetic HUD System Guides Compression Anastomosis Surgery Read More »
Picture a pathologist at a teaching hospital who trained a model last year to tell invasive ductal from invasive lobular breast carcinoma. This year the lab wants it to also separate papillary from clear cell kidney cancer, and next year…
Beyond Amnesia: Why Continual Learning is the Next Frontier for AI Pathology Read More »
Before an oncologist chooses a cancer drug, they are making an educated guess. Will this patient’s tumour respond, or will the treatment burn precious weeks while the cancer keeps growing? The truest answer would come from watching, inside the patient’s…
XPert: A Transformer That Predicts How Drugs Change Cells Read More »
Antibiotic resistance is quietly winning. The drugs that once reliably cleared an infection are failing more often, and the pipeline of new ones has thinned to a trickle. One promising class, peptide antibiotics, are short chains of amino acids that…
ApexGO Optimizes Peptide Antibiotics With Generative AI Read More »