Face Mask Detection Using Deep Hybrid Network Architectures

As the world has been severely affected by Novel Coronavirus, scientists have been working hard to study this rapidly evolving virus, its long-term and short-term implications, and how to stop its spread. As newer variants of the virus are discovered, it has become even more important to enforce the various steps required to curb its spread. We can only fight this virus by wearing masks, using sanitizers, and social distancing. This paper proposes a hybrid masked face detection model for implementing the proper use of face masks. Our study focuses on combining machine learning models and Neural Networks. Even though various models have been proposed in the past for face mask detection, we tried to change the conventional machine learning methods by creating hybrid models like ResNet50 and VGG16 and combining classical machine learning models like SVM and Gradient Booster, and Neural Networks and comparing their performance. The Hybrid model architecture consisting of ResNet50 + SVM significantly outperformed the other models, returning an accuracy and precision of more than 97 and close to 100% each respectively. © 2022, Springer Nature Switzerland AG.

A snapshot for peritoneal dialysis clinic visits: Addressing hospitalization rates with a checklist

Background: Readmission rates are a component of quality metrics in home dialysis follow-up. Common causes of peritoneal dialysis (PD) related hospitalizations have been elucidated through National Readmission Database review. However, a systematic approach to identify individual risk factors leading to the index hospitalization and targeted interventions are not directly designed into clinic workflow. Often information regarding these specific risk factors are not exacted. We identified a need to standardize practice in our PD clinic by conceiving an action checklist for nephrologists and nurses to minimize index admissions. Methods: Our quality improvement project sought to identify risk factors by analyzing the cause of admission from our cohort of 103 PD patients over 8 months. We divided reasons for admission into related and unrelated to PD. Based on these categories, we created a list of potential contributory risk-factors for admission. We also surveyed providers to determine key clinical components for a clinic checklist to encourage early recognition of the risk-factors. Results: Of the 105 individual admission events identified from June 2018 to March 2019, 45% were identified as PD-related. Such admissions included peritonitis (34%), hypervolemia (19%), electrolyte derangement (13%), hypotension (13%), hypertension (10.6%) and catheter dysfunction (10.6%). 37 admissions (35%) were readmissions in the last 30 days, of which 60% were PD-related. From these results we designed a snapshot of trends of the prior 3 months' vital signs, electrolytes, weights, PET results, PD adequacy results, urine volume, peritonitis history and current medications for clinicians to review pre-visit. Conclusions: We are currently implementing this checklist in our monthly PD clinic visits. Though the idea was conceived prior to the pandemic, we have increasingly seen the benefit of a clinical trends snapshot readily available as we transition to Telehealth visits to prevent patients' exposure to COVID-19. This method assists the clinician in triaging remotely. Ultimately, through utilization of this tool, we hope to unify our practice pattern in the clinic to reduce admission rates by prompting proactive, not reactive, interventions.

Remote peritoneal dialysis training in a COVID-19-positive patient

Introduction: Training patients in peritoneal dialysis (PD) traditionally requires up to fourteen in-person clinic visits to cover all aspects of care. The COVID-19 crisis has created an unprecedented challenge in educating patients to perform PD safely while minimizing exposure to staff. Telemedicine has been well-received by staff and patients in other aspects of PD care. We present a case of a COVID-19 positive patient who was fully trained in PD using telemedicine. Case Description: The patient is a 21-year-old man with VATER Syndrome who progressed to ESRD with uremic symptoms. He chose PD as his dialysis modality while awaiting a kidney transplant. Prior to his PD catheter insertion, he tested positive for COVID-19. He was deemed an ideal candidate for PD training via telemedicine and agreed to proceed. For the first two training sessions, he presented to the PD clinic and was placed in a designated isolation room with his personal computer. His PD nurse was in an adjoining room and trained him via video conferencing with the option to enter his room if needed. The patient quickly mastered the procedure in this monitored environment. He completed the remainder of the required education remotely in his home via telemedicine. Currently, he is fully trained and has initiated his full PD prescription. Discussion: There are several advantages of telehealth in COVID-19 patients. The risk of viral exposure to healthcare staff and other patients is reduced by limiting trips to the PD clinic. Additionally, reducing the burden of travel saves time and expense for the patient. Patient selection for telehealth learning is critical: the ideal patient must be motivated and technologically savvy. The patient, PD nurse, and nephrologist must jointly determine whether proceeding with tele-learning is feasible and safe. Although remote videoconferencing is not the conventional or optimal method for PD education, it can be used successfully to train patients while minimizing exposure of COVID-19 to staff.