Learning from small data: Classifying sex from retinal images via deep learning.

Deep learning (DL) techniques have seen tremendous interest in medical imaging, particularly in the use of convolutional neural networks (CNNs) for the development of automated diagnostic tools. The facility of its non-invasive acquisition makes retinal fundus imaging particularly amenable to such automated approaches. Recent work in the analysis of fundus images using CNNs relies on access to massive datasets for training and validation, composed of hundreds of thousands of images. However, data residency and data privacy restrictions stymie the applicability of this approach in medical settings where patient confidentiality is a mandate. Here, we showcase results for the performance of DL on small datasets to classify patient sex from fundus images-a trait thought not to be present or quantifiable in fundus images until recently. Specifically, we fine-tune a Resnet-152 model whose last layer has been modified to a fully-connected layer for binary classification. We carried out several experiments to assess performance in the small dataset context using one private (DOVS) and one public (ODIR) data source. Our models, developed using approximately 2500 fundus images, achieved test AUC scores of up to 0.72 (95% CI: [0.67, 0.77]). This corresponds to a mere 25% decrease in performance despite a nearly 1000-fold decrease in the dataset size compared to prior results in the literature. Our results show that binary classification, even with a hard task such as sex categorization from retinal fundus images, is possible with very small datasets. Our domain adaptation results show that models trained with one distribution of images may generalize well to an independent external source, as in the case of models trained on DOVS and tested on ODIR. Our results also show that eliminating poor quality images may hamper training of the CNN due to reducing the already small dataset size even further. Nevertheless, using high quality images may be an important factor as evidenced by superior generalizability of results in the domain adaptation experiments. Finally, our work shows that ensembling is an important tool in maximizing performance of deep CNNs in the context of small development datasets.

Enrichment map profiling of the cancer invasion front suggests regulation of colorectal cancer progression by the bone morphogenetic protein antagonist, gremlin-1.

The cancer invasion front (CIF), a spatially-recognized area due to the frequent presence of peritumoral desmoplastic reaction, represents a cancer site where many hallmarks of cancer metastasis occur. It is now strongly suggested that the desmoplastic microenvironment holds crucial information for determining tumor development and progression. Despite extensive research on tumor-host cell interactions at CIFs, the exact paracrine molecular network that is hardwired into the proteome of the stromal and cancer subpopulations remains partially understood. Here, we interrogated the signaling pathways and the molecular functional signatures across the proteome of a desmoplastic coculture model system of colorectal cancer progression. We discovered a group of bone morphogenetic protein (BMP) antagonists that coordinates major biological programs in CIFs, including cell proliferation, invasion, migration and differentiation processes. Using a mathematical model of cancer cell progression, coupled to in vitro cell migration assays, we demonstrated that the prominent BMP antagonist gremlin-1 (GREM1) may trigger motility of cancer cell cohorts. Our data collectively demonstrate that the desmoplastic CIFs deploy a microenvironmental signature, based on BMP antagonism, in order to regulate the motogenic fates of cancer cell cohorts invading the adjacent stroma.