COVID-19 Disease Classification by Cough Records Analysis using Machine Learning

The rapid spreading rate of the Coronavirus disease 2019 (COVID-19) has resulted in more than 6.2 million deceased cases. Furthermore, the patients of the latest Omicron variation carry light to almost no symptoms of the disease themselves. Thus, the requirement for a new diagnosis method besides Reverse Transcription-Polymerase Chain Reaction (RT-PCR) becomes the most important step to successfully detect infected cases. In this research, the application of the KNN, Ensemble and SincNet models are implemented as the main models for classification diagnosis based on cough sound records of infected patients. After pre-processing steps for removing silence ranges in the audio scripts, the cough sounds are augmented, subsequently separated into single cough samples, then generated 3 testing scenarios for dealing with the imbalanced problem between the sample classes. Afterward, MelFrequency information and MelSprectrogram are extracted as main features for analysis in order to distinguish patients with COVID-19 disease and healthy cases. The AICV115M dataset consisting of two classes COVID-19 and NonCOVID-19 is implemented for performance evaluation. The recorded highest accuracy on the models KNN, Ensemble and SincNet are 92.49%, 90.1% and 85.15%, respectively. © 2022 IEEE.

FINANCIAL IMPACT OF RACIAL DISPARITIES IN POTENTIALLY AVOIDABLE HOSPITALIZATIONS DURING THE COVID-19 PANDEMIC

BACKGROUND: Potentially avoidable hospitalizations expose patients to unnecessary iatrogenic harm, undue financial burden, and emotional stress.We previously have published that during the first 6 months of the COVID-19 pandemic, potentially avoidable hospitalizations fell by 50.3% among non- Hispanic White patients, compared to only 8.0% among African American patients at a large urban health system. Understanding the financial ramifications of this disparity is an important part of designing health policy to redress the downstream impacts of COVID-19-related healthcare inequities. METHODS: This pre-post study included 904 potentially avoidable hospitalizations (defined per the Agency for Healthcare Research and Quality guidelines) at a large urban health system between March 1 - August 31 of 2019 (pre- COVID period) and March 1 - August 31 of 2020 (COVID period). Excess healthcare expenditures were estimated from the difference in cost of potentially avoidable hospitalizations between non-Hispanic White and African American patients using hospital financial data. Lost productivity was calculated using the human capital approach by estimating the indirect cost of absenteeism from patient-specific length of stay and county wage data. County wage data was obtained from the 2020 Labor Force Statistics (U.S. Bureau of Labor Statistics) and the 2019 American Community Survey (U.S. Census Bureau). RESULTS: While African American patients experienced only a modest reduction in potentially avoidable hospitalizations (8.0%), if they had experienced the same reduction as non-Hispanic White patients (50.3%), expenditures would have been reduced by $6,587,669 during the first 6- months of the COVID-19 pandemic within this single health system. Expanding this calculation to include other minoritized groups (Asian and Latinx patients) yielded $10,465,551 in lost healthcare savings over 6 months. Non-Hispanic White patients experienced a 22.6% reduction ($111,930 to $86,601) in foregone wages, whereas African American patients experienced an increase in foregone wages of 39.5% ($24,460 to $34,113) despite having fewer hospitalizations over this time period. CONCLUSIONS: If racial/ethnic minority patients experienced comparable reductions in potentially avoidable hospitalizations as non-Hispanic White patients, expenditures at this large urban health system would have been reduced by $10.46 million during the first 6-months of the COVID-19 pandemic. Additionally, we found that the financial harms of forgone wages disproportionately burdened African American patients compared to non- Hispanic White patients. While financial cost is not the only outcome of interest when examining avoidable admissions, these findings further inform the need to develop interventions and policies to prevent avoidable admissions by improving outpatient and self- care in order to combat these disparities. Future research is needed to ascertain whether some of these reductions in avoidable admissions may have been harmful.

Apply Deep Learning on Chest X-Rays Images for COVID-19 Disease Detection by Using Transfer Learning

The outbreak of Coronavirus has caused a million fatal cases recorded globally. The challenge in dealing with the SARS-CoV-2 virus is due to its patients carrying similar symptoms with common viral pneumonia. Therefore, it is essential for doctors to recognize and differentiate the infected patients of this virus in early diagnostic steps, such as using Chest X-Ray images. For that purpose, applying transfer learning with pre-trained models is considered in this work, with the aim to single out the Corona infected images from healthy lungs or other common viral pneumonia. The Curated Dataset for COVID-19 Posterior-Anterior Chest Radiography Images (X-Rays) has been applied to train and evaluate the performance of the implemented models. The dataset consists of 4 classes with a total number of thousands of images, being Normal, COVID-19, Viral - Pneumonia, and Bacterial - Pneumonia, respectively. The high accuracy recorded results from the dataset help to nominate the suitable models for early recognition of Corona infected patients, which allows early intervention and the possibility of being completely cured of the deadly virus. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Characterization of breast cancer management during the COVID 19 pandemic in a large integrated healthcare delivery system: Stage at diagnosis and timing/modality of first treatment

Background: The COVID 19 pandemic has disrupted all aspects of healthcare, including the diagnosis and treatment of breast cancer. In March 2020, the Society of Surgical Oncology, the American College of Surgeons, and the American Society of Breast Surgeons issued guidelines regarding the timing of surgery for cancer patients to preserve hospital resources and minimize exposure of patients and staff to COVID 19. Recommendations included delaying breast cancer surgery if possible, and using neoadjuvant chemotherapy or neoadjuvant endocrine therapy to treat selected patients while waiting for definitive surgery. In California, the ?shelter in place? (SiP) order began March 17, 2020, and both screening mammography and elective surgeries were stopped in a large, integrated health care system. We evaluated the impact of these operational changes on the presentation and treatment of breast cancer patients in our system.Methods:We performed a retrospective review of patients newly diagnosed between 3/17/20, the starting date of SiP, and 5/18/20, when elective operations resumed in our system. We compared this cohort to patients who were diagnosed between 3/17/19 and 5/18/19. Age, histology, anatomic staging features, grade, receptor status, and initial treatment were compared between the cohorts. For the patients who underwent surgery, we compared the time from biopsy to time to surgery (TTS) and the type of operation. Comparisons involving categorical variables were performed using the chi-square test. Normally-distributed continuous variables were compared using two sample-t-tests. P-values of <0.05 were considered statistically significant.Results:There were 790 patients in the 2019 cohort and 279 in the 2020 cohort. There were no significant differences in age at presentation, histologic subtypes, nodal status, or operation type between the two groups. The T-stages at presentation of the 2020 group were higher than those of the 2019 group;29% presented with T1c tumors in 2020 vs 26% in 2019, and 37% with T2 tumors in 2020 vs 30% in 2019 (p=0.03). A higher percentage of patients presented with distant metastatic disease at the time of diagnosis in 2020 (7% in 2020 vs 2% in 2019, p<0.001), although the absolute numbers of patients were similar (19 patients in 2020 vs 17 patients in 2019). Of patients with invasive breast cancer, a higher percentage of patients presented with grade 3 tumors in 2020 (35% in 2020 vs 24% in 2019, p=0.002), and triple negative tumors (15% vs 10%, p=0.02). Fewer patients underwent surgery first in 2020 (73% in 2020 vs 85% in 2019, p<0.001) and more underwent neoadjuvant chemotherapy (13% in 2020 vs 9% in 2019, p=0.03). Only 4% of the 2020 surgery group had been placed on neoadjuvant endocrine therapy while awaiting definitive surgery. The TTS for patients with surgery as the initial treatment was significantly shorter for the 2020 group (mean 22 days in 2020 vs 31 days in 2019, p<0.001).Conclusions: Without screening mammography, newly-diagnosed patients in a large, integrated health care system during the COVID 19 pandemic presented with more advanced and aggressive breast cancers as compared to the equivalent time period in 2019. Fewer patients underwent surgery first, and more underwent neoadjuvant chemotherapy. The TTS for breast cancer patients in 2020 was significantly shorter than in 2019, which we hypothesize was due to the availability of operating rooms since elective operations had been stopped. This study demonstrates the ability of a large, integrated health care system to deliver timely breast cancer care to patients presenting with symptomatic disease during the constraints of the COVID 19 pandemic, and highlights the importance of screening in the early detection of breast cancer.