👤 Lourens J P Nonkes

Quantifying lymphocyte vacuolization in peripheral blood smears (PBSs) serves as a measure for disease severity in CLN3 disease-a lysosomal storage disorder of childhood-onset. However, thus far quantification methods are based on labor-intensive manual assessment of PBSs. As machine learning techniques like convolutional neural networks (CNNs) have been deployed quite successfully in detecting pathological features in PBSs, we explored whether these techniques could be utilized to automate quantification of lymphocyte vacuolization. Here, we present and validate a deep learning pipeline that automates quantification of lymphocyte vacuolization. By using two CNNs in succession, trained for cytoplasm-segmentation and vacuolization-detection, respectively, we obtained an excellent correlation with manual quantification of lymphocyte vacuolization ( Show less

The CLN3 disease spectrum ranges from a childhood-onset neurodegenerative disorder to a retina-only disease. Given the lack of metabolic disease severity markers, it may be difficult to provide adequate counseling, particularly when novel genetic variants are identified. In this study, we assessed whether lymphocyte vacuolization, a well-known yet poorly explored characteristic of CLN3 disease, could serve as a measure of disease severity. Peripheral blood obtained from healthy controls and CLN3 disease patients was used to assess lymphocyte vacuolization by (a) calculating the degree of vacuolization using light microscopy and (b) quantifying expression of lysosomal-associated membrane protein 1 (LAMP-1), using flow cytometry in lymphocyte subsets as well as a qualitative analysis using electron microscopy and ImageStream analysis. Quantifying lymphocyte vacuolization allowed to differentiate between CLN3 disease phenotypes ( Lymphocyte vacuolization serves as a proxy for CLN3 disease severity. Quantifying vacuolization may help interpretation of novel genetic variants and provide an individualized readout for upcoming therapies. Show less