During the COVID-19 pandemic, contact tracing apps based on the Bluetooth Low Energy (BLE) technology found in smartphones have been deployed by multiple countries despite BLE’s debatable performance for determining close contacts among users. Current solutions estimate proximity based on a single feature: the mean attenuation of the BLE signal. In this context, a new generation of these apps which better exploits data from the BLE signal and other sensors available on phones can be fostered. Collected data can be used to extract multiple features that feed machine learning models which can potentially improve the accuracy of today’s solutions. In this work, we consider the use of machine learning models to evaluate different feature sets that can be extracted from the received BLE signal, and assess the performance gain as more features are introduced in these models. Since indoor conditions have a strong impact in assessing the risk of being exposed to the SARS-CoV-2, we analyze the environment (indoor or outdoor) role in these models, aiming at understanding the need for apps that could increase proximity accuracy if aware of its environment. Results show that a better accuracy can be obtained in outdoor locations with respect to indoor ones, and that indoor proximity estimation can benefit more from the introduction of more features with respect to the outdoor estimation case. Accuracy can be increased about 10% when multiple features are considered if the device is aware of its environment, reaching a performance of up to 83% in indoor spaces and up to 91% in outdoor ones. These results encourage future contact tracing apps to integrate this awareness not only to better assess the associated risk of a given environment but also to improve the proximity accuracy for detecting close contacts.
Studying tissue-independent components of cancer and defining pan-cancer subtypes could be addressed using tissue-specific molecular signatures if classification errors are controlled. Since PAM50 is a well-known, United States Food and Drug Administration (FDA)-approved and commercially available breast cancer signature, we applied it with uncertainty assessment to classify tumor samples from over 33 cancer types, discarded unassigned samples, and studied the emerging tumor-agnostic molecular patterns. The percentage of unassigned samples ranged between 55.5% and 86.9% in non-breast tissues, and gene set analysis suggested that the remaining samples could be grouped into two classes (named C1 and C2) regardless of the tissue. The C2 class was more dedifferentiated, more proliferative, with higher centrosome amplification, and potentially more TP53 and RB1 mutations. We identified 28 gene sets and 95 genes mainly associated with cell-cycle progression, cell-cycle checkpoints, and DNA damage that were consistently exacerbated in the C2 class. In some cancer types, the C1/C2 classification was associated with survival and drug sensitivity, and modulated the prognostic meaning of the immune infiltrate. Our results suggest that PAM50 could be repurposed for a pan-cancer context when paired with uncertainty assessment, resulting in two classes with molecular, biological, and clinical implications.
Introduction: The high incidence of prostate cancer (PCa) worldwide and the growing interest in overdiagnosis and overtreatment make the study of new markers imperative for helping us predict the presence and aggressiveness of PCa. The objective of this work is to evaluate the-2proPSA and Prostate Health Index Usefulness for the diagnosis of PCa. Material and methods: A prospective study including 101 patients with PSA levels between 3-10 ng/mL and normal digital rectal exam (DRE) was conducted between November 2013 and November 2014. All patients underwent prostate biopsy and PSA, free PSA and-2proPSA determination.-2proPSA ratio (%2proPSA) and Prostate Health Index (PHI) were also calculated. Results: Patients had a mean age of 63.7 years old. The means of PSA and free PSA ratio (%fPSA) were 6.06 ng/mL and 16%, respectively. The means of-2proPSA and %2proP-SA were 16.8% and 1.8%, respectively. Prostate volume mean was 46 cc and PSA density mean were 0.19 ng/cc. In the univariate analysis, only %fPSA and PHI showed statistical significant association with the presence of tumor in prostate biopsy, whereas %2proPSA almost reached statistical significance. In the multivariate analysis, PHI showed the best area under the curve (AUC) with a value of 0.749, followed by %fPSA (0.708) and-2proPSA (0.671). The best values for internal and external validity of each of the evaluated parameters turned out to be for PHI, with 93% sensibility and 37% specificity, 53% positive predictive value (PPV) and 88% negative predictive value (NPV). Conclusions: PHI is the parameter that allows predicting the presence of PCa more precisely for patients with normal (DRE) and PSA between 3 and 10 ng/mL.
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.