As machine learning for images becomes democratized in the Software 2.0 era, one of the serious bottlenecks is securing enough labeled data for training. This problem is especially critical in a manufacturing setting where smart factories rely on machine learning for product quality control by analyzing industrial images. Such images are typically large and may only need to be partially analyzed where only a small portion is problematic (e.g., identifying defects on a surface). Since manual labeling these images is expensive, weak supervision is an attractive alternative where the idea is to generate weak labels that are not perfect, but can be produced at scale. Data programming is a recent paradigm in this category where it uses human knowledge in the form of labeling functions and combines them into a generative model. Data programming has been successful in applications based on text or structured data and can also be applied to images usually if one can find a way to convert them into structured data. In this work, we expand the horizon of data programming by directly applying it to images without this conversion, which is a common scenario for industrial applications. We propose Inspector Gadget, an image labeling system that combines crowdsourcing, data augmentation, and data programming to produce weak labels at scale for image classification. We perform experiments on real industrial image datasets and show that Inspector Gadget obtains better performance than other weak-labeling techniques: Snuba, GOGGLES, and self-learning baselines using convolutional neural networks (CNNs) without pre-training.
Myopia is one of the risk factors for glaucoma, making accurate diagnosis of glaucoma in myopic eyes particularly important. However, diagnosis of glaucoma in myopic eyes is challenging due to the frequent associations of distorted optic disc and distorted parapapillary and macular structures. Macular vertical scan has been suggested as a useful tool to detect glaucomatous retinal nerve fiber layer loss even in highly myopic eyes. The present study was performed to develop and validate a deep learning (DL) system to detect glaucoma in myopic eyes using macular vertical optical coherence tomography (OCT) scans and compare its diagnostic power with that of circumpapillary OCT scans. The study included a training set of 1416 eyes, a validation set of 471 eyes, a test set of 471 eyes, and an external test set of 249 eyes. The ability to diagnose glaucoma in eyes with large myopic parapapillary atrophy was greater with the vertical than the circumpapillary OCT scans, with areas under the receiver operating characteristic curves of 0.976 and 0.914, respectively. These findings suggest that DL artificial intelligence based on macular vertical scans may be a promising tool for diagnosis of glaucoma in myopic eyes.
With the recent advances in surgery due to developments in medicine, science and technology, the use of complex and enhanced equipment has improved patient care and safety. However, the use of excess anaesthetic gas, radiation, electrical and electronic equipment in the operating room, and the risk of infection and drug addiction are increasing, threatening the safety of patients and perioperative healthcare workers (Hey & Turner, 2016;Edozien, 2005). Therefore, it is ideal that perioperative nurses be sensitive to environmental factors along with awareness of patient safety and carry out preventive activities by acquiring updated knowledge of patient safety risk factors.Concerning patients' safety, risk assessment behaviours and accident types have been identified by looking at human factors (e.g. communication and behaviour according to the patient's situation; Carbral
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.