More than half of pregnant women in the UK with gestational diabetes are unaware of their condition due to insufficient diagnostics, a study suggests. This could lead to unneccessary complications.
Can an AI determine whether or not a person drinks beer by looking at their knee X-rays? It can't – but the claim shows why “shortcut learning” is such a dangerous mechanism in medical AI.
Sakura Finetek Europe and MEDIPATH, a group of independent French pathologists, announced their collaboration to enhance cancer diagnostics and thus contribute to improved patient therapeutic management. This decision is part of an ongoing commitment to improving the quality of care provided to patients and in response to the significant increase in cancer cases over the last years.
New study results could help detect whether a newborn baby with fever has sepsis or another serious bacterial infection, without the need for invasive testing.
In the world of theatre, the ‘deus ex machina’, the god from the machine, is a dramaturgical trick to resolve seemingly unsolvable conflicts. Can artificial intelligence (AI) also be such a universal problem solver for internal medicine? At the Annual Congress of the German Society of Internal Medicine (DGIM), Dr Isabella Wiest explored the potential – and limitations – of AI helpers.
When an AI advisor points out an area of concern in a chest X-ray, radiologists are sometimes all too eager to follow their lead, a new study finds. This may lead to incorrect diagnostic decisions.
Medtronic announced it has received CE Mark approval for its Endoflip 300 impedance planimetry system to measure pressure and dimensions in the oesophagus and pylorus in adults.
A new imaging technique reduces skin tone bias in breast cancer detection, improving visibility across diverse skin tones.
Early and precise diagnosis of endometriosis is crucial for women's health and quality of life. Researchers are using hyperspectral imaging and AI to improve medical care for those affected.
Not every spine abnormality shown in an MRI scan causes pain for a patient. To diagnose correctly, researchers advocate the use of questionnaires to match the images with reported symptoms.
A new method could open new avenues to quick, affordable cancer diagnosis, US researchers report. The method uses ultrathin membranes to capture extracellular vesicles.
Insulinomas are benign tumors in the pancreas, but can still cause problems for patients. A new imaging technique reliably detects these previously hard-to-find tumors, according to recent research.
Certain early-stage changes to the prostate very rarely develop into aggressive cancer, but are still called that. Would it create a false sense of security to not call these “cancer” at all?
Before administering certain drugs, doctors check a patient’s kidney function by testing their blood urea nitrogen and creatinine levels. New research shows that gold nanoparticles might give more accurate results.
Researchers have identified biomarkers that, in conjunction with PET imaging, enable doctors to reliably distinguish between primary 4-repeat tauopathies and Alzheimer's disease.
An AI-enabled digital stethoscope could help doctors identify twice as many cases of heart failure during pregnancy compared to regular obstetric care and screening, acording to a new study.
Certain long-term health conditions such as chronic obstructive pulmonary disease (COPD) and asthma, can obscure early symptoms of lung cancer, leading to delays in diagnosis, a new study finds.
Endometriosis can cause abdominal pain and cramping before, during and after menstruation, and may lead to infertility. Now, researchers identified promising approaches for better early diagnosis.
Enter symptoms into ChatGPT, receive an accurate diagnosis? Research reveals that LLM AI models are not quite there yet, struggling to identify genetic conditions from patient-written descriptions.
Using conventional ultrasound, researchers have developed a model that determines whether conditions such as fibrosis or edema are present in the lungs, and gives information about the severity.
Combining risk markers, systematic evaluation of MRI images and AI, researchers aim to predict the risk of prostate cancer more accurately than before. This could save many patients from a biopsy.
The interpretation of prostate MRI is notoriously difficult. Annotating AI shows promise to help improve diagnostic performance through increased cancer detection rates with fewer false positives.
A new type of contrast agent for use in both MRI and PET imaging has the potential to significantly enhance diagnosis and subsequent treatment, particularly for kidney diseases and tumors.
AI models perform increasingly well in answering medical questions. However, when it comes to explaining these answers, shortcomings become apparent, a new study shows.
Doctors may soon have a new, high-tech way to diagnose Multiple Sclerosis using special eye scans. This could make it easier and faster for people to find out if they have MS.