Context

Most of the time, accurate diagnosis of cancer requires a histological examination. It consists of analyzing a tissue sample so as to confirm the presence of a tumor, qualify the type of lesion detected and, thus, adjust the therapy.

At primaa, we have developed deep learning based methods to classify lesions, detect biomarkers and segment regions of interests. Those models are trained on labeled data that can be noisy. We want to explore various training strategies designed to overcome this limitation [1].

[1] :Deep learning with noisy labels: Exploring techniques and remedies in medical image analysis, Karimi et al. Medical Image Analysis, 2020,https://doi.org/10.1016/j.media.2020.101759

Objectives

• Perform a literature review on relevant approaches to deal with noisy labels.

• Select and implement relevant approaches on our various applications

• explore possible improvements

Practical information

• Remuneration : 1000€/month + 50% navigo refund

• Primaa is based in Paris, 2ème arrondissement

• possible occasional remote

• Applied mathematics student. Bachelor or Master level.

• Six months internship (start between September and October 2023.)

• Knowledge in python.

• Knowledge in computer vision (image processing, deep learning) : tensorflow and/or pytorch, scikit-image ….

Meeting with two data team members