In 2020, there were 430 000 new cases of kidney cancer, making it the 14th most common cancer worldwide and its incidence rates are rising. Renal cell carcinoma accounts for 85% of kidney cancer cases. The 5-year overall survival rate is 64% and decreases when the tumour is at a higher stage at diagnosis. Diagnosing the disease is difficult due to asymptomatic cases and non-specific symptoms. BloodCounts! seeks to create algorithms that use population-scale high-level and rich full blood count data to identify individuals at high risk of having renal cell carcinoma. This could reduce the high number of incidental findings and improve patient outcomes.
To gain a deeper understanding of renal cell carcinoma, explainability techniques will be used on the created algorithms. This will give us insights into the biological changes that happen in the blood of patients with renal cell carcinoma and it will tell us how the model uses these changes to predict clinical outcomes.