Most human activity recognition datasets that are publicly available have data captured by using either smartphones or smartwatches, which are usually placed on the waist or the wrist, respectively. These devices obtain one set of acceleration and angular velocity in the
x
-,
y
-, and
z
-axis from the accelerometer and the gyroscope planted in these devices. The PLHI-MC10 dataset contains data obtained by using 3 BioStamp nPoint® sensors from 7 physically healthy adult test subjects performing different exercise activities. These sensors are the state-of-the-art biomedical sensors manufactured by MC10. Each of the three sensors was attached to the subject externally on three muscles-Extensor Digitorum (Posterior Forearm), Gastrocnemius (Calf), and Pectoralis (Chest)-giving us three sets of 3 axial acceleration, two sets of 3 axial angular velocities, and 1 set of voltage values from the heart. Using three different sensors instead of a single sensor improves precision. It helps distinguish between human activities as it simultaneously captures the movement and contractions of various muscles from separate parts of the human body. Each test subject performed five activities (stairs, jogging, skipping, lifting kettlebell, basketball throws) in a supervised environment. The data is cleaned, filtered, and synced.
Deep learning approaches applied to medical imaging have reached near-human or betterthan-human performance on many diagnostic tasks. For instance, the CheXpert competition on detecting pathologies in chest x-rays has shown excellent multi-class classification performance. However, training and validating deep learning models require extensive collections of images and still produce false inferences, as identified by a human-in-the-loop. In this paper, we introduce a practical approach to improve the predictions of a pre-trained model through Few-Shot Learning (FSL). After training and validating a model, a small number of false inference images are collected to retrain the model using Image Triplets -a false positive or false negative, a true positive, and a true negative. The retrained FSL model produces considerable gains in performance with only a few epochs and few images. In addition, FSL opens rapid retraining opportunities for human-in-the-loop systems, where a radiologist can relabel false inferences, and the model can be quickly retrained. We compare our retrained model performance with existing FSL approaches in medical imaging that train and evaluate models at once.
Data and Code AvailabilityThis paper uses CheXpert Dataset Irvin et al. (2019), which is a publicly available dataset of 224,316 chest radiographs from 65,240 patients.The code used for experimentation processes described in this paper -which include triplet dataset creation, training and validation of the Few-Shot Learning and Incremental Few-Shot Learning Models were anonymized and included in the supplemental material section.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.