PATH News Network Daily Edition

Getting Blood Drawn By a Robot

College of American Pathologists

Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.

0:00 | 6:13
SPEAKER_01

A win in the Louisiana legislature. Getting blood drawn by a robot. And how to think about AI for pathology. These stories and more are coming up next. This is Path News Network Daily Edition from the College of American Pathologists. I'm Nick Lonnie. It's Tuesday, June 2nd. Cytotechnologists will not be recognized as cytologists in Louisiana as the legislature failed to advance a bill that would have made that name change. The CAP strongly opposed the bill along with the Louisiana Pathology Society and the Louisiana State Medical Society. They argued that cytotechnologist and pathologist are clearly defined and differentiated roles under federal law, while the term cytologist could be confused with a physician. Each state has different licensing rules and nomenclature, and the CAP will continue to monitor proposed changes across the country. Are robot phlebotomists around the corner? They could be, thanks to positive results from a clinical trial in the Netherlands reported in the journal Clinical Chemistry. The trial assessed both analytical accuracy and patient satisfaction of the Aleta Autonomous Robotic Phlebotomy device, which uses infrared light, ultrasound, and AI to detect veins and direct the needle. In the trial, no statistically significant differences were observed between human and automated blood draws across several measures of blood sample quality. The first stick success rate for the machine was 95%, versus a range of 80 to 97% for human phlebotomists in other studies. Adverse effects were minimal and mild. Importantly, more than half of patients reported less pain versus manual phlebotomy, and 82% of patients said they would strongly prefer or prefer the automated approach. Given that clinical laboratories have chronic staff shortages, don't be surprised if this technology shows up in your local lab in the coming years. It's not too late to apply for the CAP Foundation's Global Pathology Education Award. Pathologists working in under-resourced areas can receive a complimentary membership as an international fellow of the CAP and a CAP ebook subscription. Priority will be given to practicing pathologists in medically under-resourced countries as defined by the World Health Organization. This award is not open to residents. Go to foundation.cap.org to apply. The application deadline is June 16th. And finally, not a day goes by, it seems, without news of an AI innovation that might prove promising to pathology, with emphasis on might. Less than 15% of pathology slides are digitized today, which limits the potential for most AI applications. But even if it's speculative, what's the best way to assess a new AI for pathology concept? We asked Dr. Nick Reeder, CEO of Alpin Glow Biosciences in Seattle, a startup developing AI-enabled 3D spatial analysis of tissue samples. He's also a clinical acting instructor at the University of Washington.

SPEAKER_00

Looking at those studies that come out, they're exciting, uh, and some of them do turn into these really useful products that reach patients. Um, but uh it's only slightly more than an idea when it's an academic publication. Um, getting it all the way to a validated clinical product is a much longer process.

SPEAKER_01

In addition to the still low adoption of digital pathology, turning an AI idea into a clinical product requires solving a bunch of technical problems.

SPEAKER_00

Especially with these computational tools, one thing that I've seen quite a bit is that they can be developed in a specific environment. We'll say within one institution that has certain IT requirements and specific computers, we'll just say it works great there. And then you try and use it in another environment and it just falls apart. Um, so having that uh portability of um the product to other institutions can be a bigger challenge than originally thought. Um they're just little things that um that lead to that. It's not any uh you know big conceptual difference between um the two, the fully finished product and the initial academic publication. But that's where a lot of these um academic uh publications um get lost in translation, is taking it outside the walls of the institution where it was developed.

SPEAKER_01

Dr. Reeder said the AI products to watch are those that promise to expand a laboratory's capabilities and therefore persuade decision makers that investments in digital AI could lead to new revenue.

SPEAKER_00

The ones that excite me the most are tools that create a new ability that pathologists can't do right now, like predicting drug response or predicting patient outcomes. Uh, those things really excite me because to me they're number one, complementary to the work we already do. Number two is clinicians want this information, but we can't provide it to them. So they're actually adding value. And um they can provide additional kind of drive towards digital transformation.

SPEAKER_01

Well, that's all for today. See the show notes for more information on today's stories. And if you've got a story to share, write us at stories at cap.org. We'll be back Wednesday at 5 a.m. Eastern for another episode of the Daily Edition. I'm Nick Lonnie. Have a wonderful day.