#biopsies

Computing Kidneys

Analysing kidney biopsies using a novel approach with unsupervised machine learning algorithms accurately classifies chronic kidney disease and predicts future kidney function

Read the published research paper here

Image from work by Joonsang Lee and colleagues

Department of Computational Medicine and Bioinformatics and Department of Pathology, University of Michigan, Ann Arbor, MI, USA

Image originally published with a Creative Commons Attribution 4.0 International (CC BY 4.0)

Published in Scientific Reports, August 2023

since then, i’ve even seen friends ! we had a socially-distant picnic in the park on separate blankets. just me, my partner, and a couple we’re close with.

i feel good. i am nervous.

i want a diagnosis.

Cut for ranting about medical care (some discussion of surgical biopsies related to skin cancer) and what sure feels like misogyny to me.

However the news I got was good! There is some advancement on the liver disease and scarring, but nothing more than they would typically expect after 10 years of autoimmune hepatitis, so that’s good. The fatty liver seems to be down. There is some scarring, and they’re going to do another test but not until the summer, to see how much scarring exactly. They’re going to have me start doing my labwork every month instead of every three months and they’ve doubled my medicine. I was basically on the lowest dose you can have, so I went from 25mg to 50mg on the Imuran, which is one pill a day, so it’s really not that bad. I mean ‘no advancement’ on the scarring would be better, but I was like where are we here…am I just changing my diet or am I dying, what’s the deal? It’s a little more serious than just having to change my diet (which yes can be a thing if it was related to fatty liver, not just like 'do yoga and eat a salad’ kind of bs) but still the doctor isn’t terribly worried or concerned, clearly since they’re not doing any more major testing until the summer.

Look and Listen

There are many tools at doctors’ fingertips to diagnose non-alcoholic fatty liver disease – a very common condition – including ultrasound, MRI and CT scans, and biopsy. But each of these have drawbacks, ranging from expense and risk (particularly in biopsies) to unsuitability for patients with certain implants (MRI scanning). So researchers are exploring non-invasive alternatives such as photoacoustic tomography (PAT). PAT uses both light and sound waves to generate detailed images, interpreting ultrasound waves given off by structures when laser light is fired onto them. Previously, the technology couldn’t visualise blood vessel detail and movement, but a new iteration, 3D-PAT, overcomes those limitations to precisely reveal liver vessel volume, number, and distribution (as seen in this video) while carefully correcting for the natural movement caused by regular breathing. They used the technique to compare structural differences between obese and lean rats, and hope this approach could ultimately provide quick, safe information for doctors and patients.

Written by Anthony Lewis

Video from work by Xin Tong and colleagues

Caltech Optical Imaging Laboratory, Andrew and Peggy Cherng Department of Medical Engineering, Department of Electrical Engineering, California Institute of Technology, Pasadena, CA, USA

Video originally published with a Creative Commons Attribution 4.0 International (CC BY 4.0)

Published in Advanced Science, November 2022

AI vs Cancer

Thanks to sci-fi, we associate artificial intelligence (AI) with robot uprisings that enslave humanity. But – for now – AI is more likely to help than harm us. Researchers have developed a new AI algorithm to help diagnose colorectal cancer, the third most common cancer worldwide. The examination of colorectal biopsies by expert pathologists is the best way to diagnose cancer, but these specialists are overworked, leading to fatigue and risking mistakes. An algorithm, however, is only limited by its hardware. The AI, trained on clinical biopsies annotated by experts, first identifies healthy and cancerous areas and then classifies each as ‘high risk’ (red in this biopsy section) or ‘low risk’ (green). The AI classified nearly 4 in 5 biopsies correctly, and most of the mistakes were false positives, meaning that it's unlikely to miss cancer. After further improvements, the aim is to introduce the AI as a tool to assist pathologists soon.

Written by Henry Stennett

Image from work by Cowan Ho, Zitong Zhao and Xiu Fen Chen, and colleagues

Department of Anatomical Pathology, Singapore General Hospital, Singapore

Image originally published with a Creative Commons Attribution 4.0 International (CC BY 4.0)

Published in Scientific Reports, February 2022

Today's biopsy went better than yesterday's, and radiology thinks the mass they biopsied today is a fibroadenoma.

Still nervous about yesterday's biopsies. Twin Bro is also nervous. When an MD whose experience is in women's health is nervous, I'm hella nervous. I think I heard him cringe over the phone.

And surgery is still on for tomorrow...