Deep neural networks have shown promising results in disease detection and classification using medical image data. However, they still suffer from the challenges of handling real-world scenarios especially reliably detecting out-of-distribution (OoD) samples. We propose an approach to robustly classify OoD samples in skin and malaria images without the need to access labeled OoD samples during training. Specifically, we use metric learning along with logistic regression to force the deep networks to learn much rich class representative features. To guide the learning process against the OoD examples, we generate ID similar-looking examples by either removing class-specific salient regions in the image or permuting image parts and distancing them away from in-distribution samples. During inference time, the K-reciprocal nearest neighbor is employed to detect out-of-distribution samples. For skin cancer OoD detection, we employ two standard benchmark skin cancer ISIC datasets as ID, and six different datasets with varying difficulty levels were taken as out of distribution. For malaria OoD detection, we use the BBBC041 malaria dataset as ID and five different challenging datasets as out of distribution. We achieved state-of-the-art results, improving 5% and 4% in TNR@ TPR95% over the previous state-of-the-art for skin cancer and malaria OoD detection respectively.
Background and Objective: Hepatitis B virus (HBV) infection is one of the common chronic viral infections worldwide. The World Health Organization (WHO) had estimated that only a minority are aware of their status and still a fraction of the diagnosed cases were receiving treatment for their ailment. Theaverageburden of disease in Pakistan is about 3.3%. The incidence may appear lower among the healthcare workers (HCWs) as compared to blood donors, but HCWs are at increased risk to acquire the HBV infection and the risk is even greater for laboratory personnel. This risk can be minimized by offering them the vaccination.Methods: Blood samples from all categories of staff in the laboratory were tested for hepatitis B Surface Antibodies (HBsAb). Those foundnon-immune were offered recombinant hepatitis B vaccine by intramuscular (I/M) injections. The results were tabulated by entering age/sex of the laboratory staff with HBsAb level. Blood was drawn 4 weeks after completion of 3 doses vaccination course in 6 months. Repeat HBsAb levels were determined in this initially non-immune group. All results were analyzed using SPSS version 21.Results: Out of 96 staff members, 30were found to be immune on first testing. Remaining 66 non-immune staff were offered complete course (3 doses) of HBV Vaccine. Five staff members were lost to follow up during the course of vaccination, two of them refused vaccination and two were non-responders. The rest 57 became immune after three doses of vaccination. Conclusion:It is recommended that a national policy be adopted for HBsAb screening and offer of vaccination, to non-immune HCWs.
Background and Objective: Platelet transfusion is one of the most crucial therapeutic approaches in medicine. Single Donor Platelets (SDP) are being preferred because of higher platelet count per unit, leukocyte reduction during collection and fewer donor exposures thus reducing the risk of infection and alloimmunization. This study was conducted to compare Haemonetics MCS plus with Baxter CS3000 plus cell separator in terms of processing time, quality of platelet concentrates, donor experience and individual choice. Methods: Two hundred platelet pheresis procedures performed on Haemonetics MCS plus during a period from January 2018 to August 2019 were compared with the same number of procedures performed using Baxter CS3000 plus cell separator from July 2015 to April 2019. Results: The mean platelet count of the product was higher with Baxter, 1741.6 ± 347 x 10 3 /μL as compared to1676 ± 301 x 10 3 /μL with Haemonetics. No significant difference between the two instruments was observed regarding processing time, product volume and yield. Conclusion:The two instruments are comparable in terms of time, volume and yield of the product but Haemonetics is better because of donors' comfort and for being operator friendly.
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.