DeepHeme is a ResNeXt-50 architecture.
Input is a local magnified image of a cell in the centre.
Output is a classification of the cell into the 23 classes — each being a type of cell. I’ll think about later what do these classes mean.
It falls under automated
Practical applications of
The AI can be much faster and more accuracy.
50 slides of BMAs from UCSF Parnassus (adult) and UCSF Benioff (children) from 2017 to 2020.
Slides show “
uninvolved marrow and
Each slide contains around 10,000 cell images that are patched out, and labelled by the “consensus decision of an expert panel of three hematopathologists”.
The object detection model
DeepHeme is then trained via
e.g. self-supervised representation learning, latent-space clustering, GANs, …
The 23 cell types include:
- All those being looked at in a standard bone marrow differential
- The ofdifferentiation stagestrilineage hematopoietic cells
- See the picture for details (I counted, there are 23 of them!)
- andartifactsmitotic bodies
The model is being tested on a held-out validation set, as well as datasets from different hospitals such as MSKCC, UCSF and Brigham and Women’s Hospital), and competed against human experts to achieve faster, more accurate and precise results.
For the mathematics, I will need to know more about
To understand the implications of the UMAP results for blood cell morphology, I’m going to need to read
- The BMA cell images used in this paper come from BMAs that exhibit quite normal . The model is expected to perform poorly for cells with more abnormal and deviant morphology. We need to use a much more diverse dataset that includes different kinds of pathological and abnormal BMA.hematopoiesis
- Though labelled cell images are hard to come by, we might have a lot more unlabelled BMAs than labelled ones. We could potentially use techniques beyond supervised learning to improve the downstream supervised learning results.
- As a thought toward practical model generalizability, since the current DeepHeme model is so high-performing, have we considered the potential incorporation of conformal prediction, as a method of model performance checking, as well as a kind of model adaptation with theoretical performance guarantee?