David S. Smith

Tahsin Reasat, Asif Sushmit, David S. Smith

Deep learning (DL) based diagnostics systems can provide accurate and robust quantitative analysis in digital pathology. These algorithms require large amounts of annotated training data which is impractical in pathology due to the high resolution of histopathological images. Hence, self-supervised methods have been proposed to learn features using ad-hoc pretext tasks. The self-supervised training process uses a large unlabeled dataset which makes the learning process time consuming. In this work, we propose a new method for actively sampling informative members from the training set using a small proxy network, decreasing sample requirement by 93% and training time by 62% while maintaining the same performance of the traditional self-supervised learning method. The code is available on https://github.com/Reasat/data_efficient_cl

Tahsin Reasat, Stephen Chenard, Akhil Rekulapelli et al.

Accurate musculoskeletal soft tissue tumor segmentation is vital for assessing tumor size, location, diagnosis, and response to treatment, thereby influencing patient outcomes. However, segmentation of these tumors requires clinical expertise, and an automated segmentation model would save valuable time for both clinician and patient. Training an automatic model requires a large dataset of annotated images. In this work, we describe the collection of an MR imaging dataset of 199 musculoskeletal soft tissue tumors from 199 patients. We trained segmentation models on this dataset and then benchmarked them on a publicly available dataset. Our model achieved the state-of-the-art dice score of 0.79 out of the box without any fine tuning, which shows the diversity and utility of our curated dataset. We analyzed the model predictions and found that its performance suffered on fibrous and vascular tumors due to their diverse anatomical location, size, and intensity heterogeneity. The code and models are available in the following github repository, https://github.com/Reasat/mstt

David S. Smith

Raw data sizes are growing and proliferating in scientific research, driven by the success of data-hungry computational methods, such as machine learning. The preponderance of proprietary and shoehorned data formats make computations slower and make it harder to reproduce research and to port methods to new platforms. Here we present the RawArray format: a simple, fast, and extensible format for archival storage of multidimensional numeric arrays on disk. The RawArray file format is a simple concatenation of a header array and a data array. The header comprises seven or more 64-bit unsigned integers. The array data can be anything. Arbitrary user metadata can be appended to an RawArray file if desired, for example to store measurement details, color palettes, or geolocation data. We present benchmarks showing a factor of 2--3$\times$ speedup over HDF5 for a range of array sizes and a speedup of up to 20$\times$ in reading the common deep learning datasets MNIST and CIFAR10.