It seems logical: You notice something funny on your skin, but you don’t want to take the time to go to a doctor to see whether it’s cancerous, so you check it out instead by taking a picture with your phone and submitting it to an online app. If the app gives you the all-clear, you can quickly get on with your life, right? But a study from the University of Pittsburgh Medical Center finds that such skin-check apps are unreliable. In three out of four apps reviewed, roughly 30 percent of marks were dismissed as unconcerning, when they were really potentially fatal skin cancers, Joel A. Wolf and his colleagues found.
The authors didn’t identify the apps, but said they found one that was trustworthy: A site that sent the pictures directly to a board-certified dermatologist accurately diagnosed melanoma 98 percent of the time. The others examined skin photos with an automated algorithm. The application that involved doctors was also the most expensive at $5 per lesion analyzed, and took about 24 hours, compared with nearly instantaneous results with the automated apps.
BOTTOM LINE: Doctors are still better at diagnosing skin cancer than smartphone apps that rely on computers.
CAUTIONS: The study looked at only four online applications. The technology might improve as cellphone cameras get better.
WHERE TO FIND IT: JAMA Dermatology, online Jan. 16.
Blood clot risk higher with IVF
About 5 million people have been born via in-vitro fertilization and the procedure is generally considered safe for both babies and mothers. But a study from the Karolinksa Institute in Sweden found a higher risk of dangerous blood clots in women pregnant via in-vitro fertilization. Smaller studies had hinted at an increased risk for some clots in the lung and elsewhere, but this is the first large-scale study to confirm that danger and to show that the highest risk comes during the first trimester of pregnancy. High estrogen levels from IVF may be to blame, the researchers suggest, citing established links between blood clots and estrogen pills.
The study compared 23,498 Swedish women who were pregnant via IVF between 1990 and 2008 with 116,960 women who had unassisted pregnancies during that time. The risk of a clot is still low in women who receive IVF, but because such clots are a leading cause of maternal death during pregnancy, the researchers wanted to highlight the extra risk posed by IVF.
BOTTOM LINE: Doctors may want to consider blood thinning medication for women who get pregnant via in-vitro fertilization.
CAUTIONS: Women who get IVF receive more attention from their doctors, so the clots may be more likely to be picked up. The study did not look at the seriousness of each clot, and did not include women who had died during pregnancy.
WHERE TO FIND IT: British Medical Journal, Jan. 15.