Diabetes technology (DT) has accomplished tremendous progress in the past decades, aiming to convert these technologies as viable treatment options for the benefit of patients with diabetes (PWD). Despite the advances, PWD face multiple challenges with the efficient management of type 1 diabetes. Most of the promising and innovative technological developments are not accessible to a larger proportion of PWD. The slow pace of development and commercialization, overpricing, and lack of peer support are few such factors leading to inequitable access to the innovations in DT. Highly motivated and tech-savvy members of the diabetes community have therefore come up with the #WeAreNotWaiting movement and started developing their own do-it-yourself artificial pancreas systems (DIYAPS) integrating continuous glucose monitoring (CGM), insulin pumps, and smartphone technology to run openly shared algorithms to achieve appreciable glycemic control and quality of life (QoL). These systems use tailor-made interventions to achieve automated insulin delivery (AID) and are not commercialized or regulated. Online social network megatrends such as GitHub, CGM in the Cloud, and Twitter have been providing platforms to share these open source technologies and user experiences. Observational studies, anecdotal evidence, and realworld patient stories revealed significant improvements in time in range (TIR), time in hypoglycemia (TIHypo), HbA1c levels, and QoL after the initiation of DIYAPS. But this unregulated do-it-yourself (DIY) approach is perceived with great circumspection by healthcare professionals (HCP), regulatory bodies, and device manufacturers, making users the ultimate riskbearers. The use of the regularized CGM and insulin pump with unauthorized algorithms makes them off-label and has been a matter of great concern. Besides these, lack of safety data, funding or insurance coverage, ethical, and legal issues are roadblocks to the unanimous acceptance of these systems among patients with type 1 diabetes (T1D). A multi-agency approach is necessary to evaluate the risks, and to delineate the incumbency and liability of Digital Features To view digital features for this article go to

No abstract

No abstract

No abstract

Early detection of incipient faults is of vital importance to reducing maintenance costs, saving energy, and enhancing occupant comfort in buildings. Popular supervised learning models such as deep neural networks are considered promising due to their ability to directly learn from labeled fault data; however, it is known that the performance of supervised learning approaches highly relies on the availability and quality of labeled training data. In Fault Detection and Diagnosis (FDD) applications, the lack of labeled incipient fault data has posed a major challenge to applying these supervised learning techniques to commercial buildings. To overcome this challenge, this paper proposes using Monte Carlo dropout (MC-dropout) to enhance the supervised learning pipeline, so that the resulting neural network is able to detect and diagnose unseen incipient fault examples. We also examine the proposed MC-dropout method on the RP-1043 dataset to demonstrate its effectiveness in indicating the most likely incipient fault types.