Continuously learning AI pathologist: An AI powered smart microscope - - PowerPoint PPT Presentation

Continuously learning AI pathologist: An AI powered smart microscope that can automatically scan different biological samples

Tathagato Rai Dastidar, Co-founder & Chief Scientific Officer, SigTuple

The Wonder Tool!

The clinical microscope

Still a gold standard for detecting various types of abnormalities in clinical laboratories

… and the Wonder Person!

The clinical pathologist

But there aren’t enough ...

HOW MANY OF YOU ARE THERE? NOT ENOUGH

Some numbers: In India,

• 1.2 Billion people
• 20,000 pathologists!

The continuously learning AI pathologist

Introducing ...

Transformation over a year ...

How it works

Seeing is believing!

Enabling remote review and collaboration

Multi purpose ...

Blood Urine Semen

… and many more in the pipeline

Continuous learning

Unidentified cells

New data Re-training Extensive validation Deployment

Continuous learning

• Predictions model become better, without any

changes in the hardware

• Predictors for new disease conditions made

available

Where we are ...

• Several clinical studies with leading

laboratories

• High correlation of statistical indices

with existing state-of-the-art

• Adept at finding rarer cells which a

pathologist or existing machines typically miss ○ Proven across multiple studies

Where all does AI play a role?

Scanning

0.5mm diameter

• Impossible to capture entire slide (75mm X

25mm) ○ Location and size of “analyzable area” uncertain ○ Use case dependent definition of “good area”

• Intelligent way to figure out “promising” areas
• Machine learned models to identify “good

images” and “good movement directions”

Classification

• Images are 4k X 3k, with thousands of objects

○ Cannot be classified as a whole

• General approach: Localization followed by classification
• Performance a major bottleneck for CNN based localization

RBC WBC Platelet Normal, elliptical, fragmented ... Neutrophil, lymphocyte, monocyte ... Normal, large ...

Volume estimation

Red blood cells under a microscope What they actually look like

3D characteristic estimation from 2D images

Preventive maintenance

Machine

• peration logs

AI/ML based analysis Dear customer, your device is likely to develop motor problems in the next 48 hours ...

Thank you! Questions?