ESR8 – Three completed secondments at the UPV, Tyris and INCLIVA

Secondments are the most exciting months and suppose great opportunities to learn in different areas, meet interesting people and advance in our researches

Arne Schmidt is Early Stage Researcher (ESR) 8 of the CLARIFY project at the Universidad de Granada.

Interdisciplinary exchange is the key to the successful application of AI in digital pathology. That’s why I completed different secondments in Valencia in different areas – computer vision, software development, and pathology.

The first secondment took place at the UPV where we initiated a collaboration about Triple Negative Breast Cancer (TNBC). We want to apply probabilistic multiple instance learning algorithms to help pathologists with the estimation of important prognostic factors like the histological subtype or the estimated number of lymph node metastasis. In this context, I was closely collaborating with Claudio (ESR6, UPV).

The secondment at Tyris Software was focused on practical software engineering, interfaces, and MLOps, which includes all techniques to deploy and maintain machine learning models in production reliably and efficiently. I also had the opportunity to learn more about the scientific work of Zahra (ESR9) who works on content-based image retrieval in the CLARIFY project.

Zahra Tabatabaei (ESR9) and Arne Schmidt (ESR8) at Tyris Software.

The third secondment at INCLIVA allowed me to learn more about histopathology in practice. In the pathology laboratory, I had the chance to learn about skin, breast, and prostate cancer firsthand – from expert pathologists in this field. Furthermore, I could observe the complete preparation process of Whole Slide Images from tissue extraction to the final scanning of the biopsies. Andrés (ESR12) kindly explained the complete process and the basics of pathology like different tissue types and their function. Another important experience was accompanying the pathologists during the diagnostic process of biopsies. This helped to understand what problems the pathologists are facing – and how AI can help in practice.