Bayesian Optimization-Based Diagnosis of Covid-19 Chest X-Rays -Ai Perspective

Coronavirus is the deadliest disease globally, and no efficient treatment has been established. The prognosis of illnesses caused by virus outbreaks is a severe medical process that demands a large amount of accurate data comprised of many factors to produce an appropriate analysis. We have researched and analyzed the factors that might affect humans and increase the chances of infection with Covid-19. One of them is the breathing symptoms directly affecting the lungs and chest. To analyze the factors, we have used traditional machine learning and deep learning models to classify and predict the chances of a human getting infected with different SARs variants. So, we used a Cyclic Generative Adversarial Networks (CGANs) model, Convolutional Neural Networks (CNNs), to generate, predict and classify the Covid-19 occurrence through chest x-rays and other attributes like Diabetes and Hypertension. These models are deployed to the cloud with appropriate hypermeter tuning to use the result in real time. This paper proposed CGANs and CNNs, which automatically use ADAM, RMSprop and Bayesian optimizers to identify chest X-ray COVID-19 pneumonia images. Then, using extracted features has increased the performance of the proposed technique. The experiments suggest that the presented ADAM method fits RMSprop and Bayesian optimization achieves better accuracy. Within proposed algorithms, Bayesian optimization effectively predicts the diagnosis of covid-19 patients.Copyright All © 2023 are reserved by International Journal of Pharmaceutical Sciences and Research.

Innovative Approaches to Data Collection Needs in Clinical Trials: Considerations, Lessons Learned and Future Directions

Technological advances and unique data collection requirements across multiple data streams have led to a proliferation of options for data collection in clinical trials. This includes traditional clinical trial data collected in 21CFR Part 11-compliant electronic casereport forms, patient-reported outcomes via standardized assessments and/or computerized tasks, passive or interactive data collection from application-based systems, provider, and/or hospital/clinic-level surveys. The COVID-19 pandemic has led to additional data collection considerations and necessitated policy changes accelerating the ability to conduct health care, clinical trial study visits, and other data collection procedures remotely. Such developments will leave enduring marks on policy related to healthcare and the conduct of clinical trials. The National Institute on Drug Abuse Clinical Trials Network (NIDA CTN) conducts multisite clinical trials on pharmacological and behavioral interventions for Substance Use Disorders (SUDs). The Data and Statistics Center at Emmes is responsible for developing data systems across many NIDA CTN trials and works with Lead Investigative teams such as the New York Node to develop novel approaches to data collection that suit the needs of individuals with SUDs and the studies designed to test and implement effective treatments. This presentation will include different perspectives from members of these teams. Challenges with technological literacy, access to smartphone devices with active data plans, and limitations to remote support from research staff can be barriers to completing assessments for SUD populations. The development of different data system approaches allowing (1) remote completion of web-based surveys and (2) encrypted, automated remote login to a 21CFR11-compliant system (no user ID and password required) will be discussed, as well as interactions between systems and considerations of approaches for different clinical research scenarios. For example, populating a survey link onto an eCRF if research staff assistance is needed for survey completion or for manual distribution of unique links embedded with participant information. Pros and cons of varied approaches to allow participants to access surveys will be discussed, including QR codes, automated vs personalized distribution via SMS and email, and mailed paper invitations with vanity URLs, which may differ by respondent population. In addition, studies targeting a specific patient population may require prescreening a large number in order to attain enrollment targets (e.g. tens of thousands of primary care patients who are sub-clinical threshold but at-risk for SUD). Approaches for conducting extensive prescreening anonymously and lessons learned (predicting prescreen targets accurately, monitoring enrollment rate, staffing and resource considerations) will be presented. Finally, an increasing number of clinical trials examine implementation of SUD interventions in specific patient populations or settings alongside traditional efficacy outcomes. Implementation assessments present unique considerations for data collection;in particular, the need to collect pre-implementation assessments of provider and site-level readiness prior to the onset of training and the new intervention;and before sites have been endorsed for trial data collection in order to avoid contamination. Another unique consideration involves qualitative data collection around the implementation process, and any adaptations made to the intervention or implementation facilitation based on a combination of qualitative and quantitative data. Balancing implementation needs with efficacy testing in clinical trials will also be discussed. Dr. Kathryn Hefner is a clinical psychologist and the Scientific Director of the National Institute on Drug Abuse Clinical Trials Network's (NIDA CTN) Data and Statistics Center (DSC). Her research interests involve substance use and effective treatments for substance use and comorbid mental health conditions. She leads the DSC's efforts in p tient-reported outcomes, including the adoption of novel data collection practices for the NIDA CTN. In addition, she serves as Co- Lead Investigator on CTN-0126, a longitudinal followup study of participants in CTN-0097 and CTN-0100. Dr. Matisyahu Shulman is a clinician-scientist at New York State Psychiatric Institute (NYSPI) and Columbia University Irving Medical Center. His research focus is on opioid use disorders clinical trials and the use of technology to enhance implementation, quality improvement, and treatment delivery. He is part of the lead team of several large multisite opioid treatment trials, including the CTN-0097 trial, a hybrid type I effectiveness-implementation trial, and CTN- 0126, a longitudinal follow-up study. Ms. Onumara Opara, MPH, is the National Project Director for CTN-0097. She has over 9 years of experience in project management for clinical and community-based research. Onumara oversees the dayto- day operations of the CTN-0097 trial, including coordination of protocol activities, implementation efforts at clinical sites, organization and supervision of study teams, training of research staff, quality assurance, and monitoring study progress. Ms. Christina Scheele,MA, is a Senior Data Manager at the CTN Data and Statistics Center at the Emmes Company. She has supported multiple CTN studies during system development and helped to troubleshoot and implement novel approaches to data collection to ease participant and staff burden. Ms. Scheele also has experience using multiple electronic database capture systems for other research studies which required unique and innovative system development for implementation. She currently serves as Data Management Lead on the CTN- 0097, CTN-0100, and CTN-0126 studies. Rebecca Price is a Senior Data Manager who has been at Emmes since 2016, currently supporting the NIDA DSC. Rebecca's main role has included leading the development, maintenance, and/or closeout activities for multiple CTN studies, and she has played a central role in adopting novel data collection practices across several NIDA CTN studies. Dr. Greiner is an addiction psychiatrist in training and a T32 postdoctoral fellow in the Division of Substance Use Disorders at Columbia University and NYSPI. Her main research interest is in implementation and dissemination of evidence-based interventions for substance use disorders in community settings. In particular, she is interested in implementation strategies, costs associated with implementing and sustaining interventions across different organizations, and developing methodology around implementation process. Dr. Greiner will discuss the hybrid effectiveness-implementation NIDA CTN (CTN-0097) trial led by principal investigators Drs. Adam Bisaga and Edward Nunes.

An Efficient method for Recognition of Occluded Faces from Images

The detection of masked face is becoming an essential part of health care safetydue to the pandemic caused by the coronavirus and the surveillance systems. One of the most challenging problems in face recognition systems is the accurate identification of faces in the presence of occlusion like wearing of glasses and masks. The current study proposes a novel convolutional neural network(CNN)-based model for accurate detection of faces in the presence of mask and glasses.The novel architecture of the model was developed using ten convolutional layers, five max-pooling layers, and a dropout layer. The Adam optimizer was used for optimization of the performance our model. Early stopping criteria in conjunction with the ReduceLROnPlateau class was employed to avoid the overfitting problem. Our proposed model could achieve the accuracy of 99.71% on the test dataset suggesting its superiorityto its existing counterparts. Based on the results, the suitability of the proposed model for face detection in the presence of occlution in real-life application has been recommended. Copyright © 2022, Anka Publishers. All rights reserved.

THE CLOT THICKENS: THE CASE OF A YOUNG MALE WITH RECURRENT PULMONARY EMBOLISMS

SESSION TITLE: Rare Cases with Masquerading Pulmonary Symptoms SESSION TYPE: Rapid Fire Case Reports PRESENTED ON: 10/18/2022 01:35 pm - 02:35 pm INTRODUCTION: Paget-Schroetter (PS) syndrome, also known as venous thoracic outlet syndrome, is a primary thromboembolic sequela of compression of the subclavian vein. CASE PRESENTATION: A previously healthy 24 year old male presented with shortness of breath and cough. He had recently been exposed to COVID. He denied fever, diarrhea, vomiting, leg swelling, and rashes. On physical exam he was tachycardic, had no murmurs or JVD, and was CTAB with no wheezing or rales. Labs were significant for a WBC of 17k, and troponin of 0.033. CTA of the chest showed multiple filling defects in the pulmonary arteries consistent with Pulmonary Embolism (PE). He was started on a heparin drip. All COVID testing was negative. Lower extremity venous doppler ultrasounds (US) were negative for DVT. His respiratory status improved, and he was discharged on apixaban with the diagnosis of PE provoked by possible COVID infection. He returned approximately 2 months later with exertional dyspnea and upper extremity swelling and was found to have recurrent PE despite having been compliant with his apixaban. Upper extremity venous doppler US was significant for DVT in his right subclavian vein. He was placed on warfarin. At this time his hypercoagulable workup was also negative. Symptoms persisted despite being on warfarin with outpatient monitored INR. A venogram was ordered to evaluate upper torso blood flow. The venogram was remarkable for high-grade stenosis of the right subclavian vein. This finding led to the consideration of thoracic outlet syndrome aka Paget-Schroetter (PS). DISCUSSION: PS is a rare clinical entity that results from stress placed on the endothelium of the subclavian vein as it passes between the junction of the first rib and the clavicle. It can predispose otherwise healthy patients to recurrent venous thromboembolisms that are refractory to anticoagulation. The clinical features usually include upper extremity swelling and pain which is exacerbated by repetitive or strenuous exercise. Venous collaterals can also be seen in some patients. Evaluation should include some form of upper extremity Doppler and a CT/MR venogram or venography to make the final diagnosis. Treatment may involve anticoagulation, thrombolysis, and/or surgical decompression. Best results are seen with early thrombolysis and surgical decompression. If caught early and treated appropriately, PS has a good outcome with few long-term sequela. CONCLUSIONS: Our goal was to describe a patient with an uncommon cause for recurrent venous thromboembolisms that were refractory to anticoagulation. Our patient's presentation of PS serves to describe many aspects of the disease process, evaluation, diagnosis, and management as seen in the case presentation. The patient's demographic fit the epidemiological profile age of 20s-30s with typical imaging findings and pertinent negative workup which would lead providers to this rarer diagnosis. Reference #1: Saleem T, Baril DT. Paget Schroetter Syndrome. [Updated 2022 Jan 11]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing;2022 Jan-. https://www.ncbi.nlm.nih.gov/books/NBK482416/ Reference #2: Alla VM, Natarajan N, Kaushik M, Warrier R, Nair CK. Paget-schroetter syndrome: review of pathogenesis and treatment of effort thrombosis. West J Emerg Med. 2010;11(4):358-362. Reference #3: Karl A. Illig, Adam J. Doyle, A comprehensive review of Paget-Schroetter syndrome, Journal of Vascular Surgery, Volume 51, Issue 6, 2010,Pages 1538-1547,ISSN 0741-5214, https://doi.org/10.1016/j.jvs.2009.12.022. DISCLOSURES: No relevant relationships by Jonathan Marks No relevant relationships by Zachary Stachura

SMARCA4-DEFICIENT UNDIFFERENTIATED TUMOR: A RARE THORACIC MALIGNANCY

SESSION TITLE: Rare Malignancies SESSION TYPE: Case Reports PRESENTED ON: 10/17/2022 03:15 pm - 04:15 pm INTRODUCTION: SMARCA4 deficient undifferentiated tumors (SMARCA4-DUT) are rare and aggressive neoplasms that are most commonly encountered in young male smokers and portend a poor prognosis (1,2). They are characterized by loss of SMARCA4, a subunit of chromatin remodeling complexes, and loss of the tumor suppressor brahma-related gene 1 (BRG1). We present a case of an elderly female with an extensive smoking history who was diagnosed with SMARCA4-DUT. CASE PRESENTATION: An 84 year old female with approximately 70 pack year smoking history, emphysema, ischemic cardiomyopathy, and coronary artery disease, presented to the emergency room with upper abdominal pain which started one day prior to admission. She endorsed an unintentional 10 pound weight loss in the past two months. The patient was admitted for an incarcerated ventral hernia for which she underwent repair. Of note, one and a half years ago, she was found to have a right lower lobe 7mm nodule but was unable to follow up due to the COVID pandemic. On this admission, a CT chest revealed a 4.2 x 3.8 x 3.7cm mediastinal mass and subcarincal lymphadenopathy. She underwent an EBUS with biopsy of the mediastinal mass and subcarinal lymph node. Cytology showed highly atypical epitheloid cells, concerning for a neoplasm with neuroendocrine differentiation and granulomas. Given the high suspicion for malignancy, she had a PET CT (figure 1) which showed FDG activity (SUV 11) in the mass with areas of necrosis and was referred to thoracic surgery. She underwent thoracoscopy with mediastinal mass resection and lymph node dissection and pathology showed diffuse sheets of epithelioid cells with large foci of necrosis. Neoplastic cells showed preserved INI (SMARCB1) expression, non-reactivity for NUT, and complete loss of BRG1 (SMARCA4) expression, consistent with a SMARCA4-DUT with positive margins (figure 2). She was referred to Radiation Oncology with plans to pursue further therapy thereafter. DISCUSSION: SMARCA4-DUT is a new and distinctive clinicopathological entity of aggressive thoracic tumors (1). The novelty of this class of tumors poses challenges in terms of treatment. Immune checkpoint inhibitors have shown compelling outcomes in case reports (3), however larger studies are needed to delineate optimal treatment regimens. CONCLUSIONS: SMARCA4-DUT are are rare but highly aggressive thoracic neoplasms. They present as large tumors and are smoking related. Prompt recognition may aid in early diagnosis. No definitive therapy exists but immunotherapy has shown promising results. Reference #1: Chatzopoulos, K., Boland, J.M. Update on genetically defined lung neoplasms: NUT carcinoma and thoracic SMARCA4-deficient undifferentiated tumors. Virchows Arch 478, 21–30 (2021). Reference #2: Roden AC. Thoracic SMARCA4-deficient undifferentiated tumor-a case of an aggressive neoplasm-case report. Mediastinum. 2021;5:39. Published 2021 Dec 25. Reference #3: Henon C, Blay JY, Massard C, Mir O, Bahleda R, Dumont S, Postel-Vinay S, Adam J, Soria JC, Le Cesne A. Long lasting major response to pembrolizumab in a thoracic malignant rhabdoid-like SMARCA4-deficient tumor. Ann Oncol. 2019 Aug 1;30(8):1401-1403. DISCLOSURES: No relevant relationships by Sathya Alekhya Bukkuri No relevant relationships by Erin Meier No relevant relationships by Mangalore Amith Shenoy No relevant relationships by Alexandra Zavin

HIGH-FLOW NASAL CANULA AND INHALED EPOPROSTENOL RESPONSE IN COVID-19

SESSION TITLE: Medications and Pulmonary Rehabilitation in COVID-19 Posters SESSION TYPE: Original Investigation Posters PRESENTED ON: 10/18/2022 01:30 pm - 02:30 pm PURPOSE: The use of inhaled epoprostenol (iEPO) has demonstrated improvement in outcomes for patients with pulmonary hypertension and right heart failure. iEPO has been used as a rescue therapy for acute respiratory distress syndrome (ARDS) and has been shown to improve oxygenation, reduce shunting, and decrease pulmonary artery pressures. However, pulmonary vasodilators do not improve mortality in patients with ARDS. Furthermore, there is currently little data on the efficacy of iEPO via high flow nasal cannula (HFNC) for ARDS patients. Here, we describe our experience with iEPO in our patients with COVID-19-related ARDS on HFNC in a Northern California county hospital. METHODS: From March 2020 to December 2021, 74 patients with COVID-19 infection and related ARDS were placed on HFNC and received iEPO, at a public tertiary care center. A positive response to iEPO was defined as an increase P/F ratio of 10%, increase in PaO2 of 20%, decrease in FiO2, or reduced flow rate within 24 hours of initiation of iEPO. Non-parametric statistics were used to compare groups. RESULTS: 21 women and 53 men with COVID ARDS ranging from 30-86 years of age (mean age 60.1 ± 13.9) received iEPO while on HFNC. The mean hospital length of stay was 36.3 ± 43 days. All patients received steroids and 83.8% received antibiotics. 55.4% of all patients in the study (n=41) progressed to mechanical ventilation and 58.1% (n=43) survived to discharge, mean age 57 ± 14 years. 20.3% (n=15) of patients showed a response to iEPO. Patients who responded to iEPO were significantly less likely to progress to mechanical ventilation (13% vs 66%, p=0.0003) and more likely to survive to discharge (93% vs 49%, p=0.0021). CONCLUSIONS: Among patients with COVID ARDS on HFNC, patients who respond to iEPO are less likely to progress to mechanical ventilation and more likely to survive to discharge. Our study is limited by small sample size and lack of randomization. Use of iEPO in the right subset COVID ARDS on HFNC may improve outcomes. CLINICAL IMPLICATIONS: Patients on HFNC selected for initiation of iEPO had a poor overall prognosis, with 41.9% not surviving to discharge and 55.4% requiring mechanical ventilation. iEPO response correlates with not requiring mechanical ventilation and with increased likelihood of survival to discharge. DISCLOSURES: No relevant relationships by Heng Duong No relevant relationships by Craig Ivie No relevant relationships by Neharika Khurana No relevant relationships by Connie Park No relevant relationships by Natasha Puri No relevant relationships by Adam Thompson No relevant relationships by John Wehner

PLASMAPHERESIS FOR HYPERTRIGLYCERIDEMIA-INDUCED PANCREATITIS

SESSION TITLE: Critical Renal and Endocrine Disorders Case Report Posters SESSION TYPE: Case Report Posters PRESENTED ON: 10/17/2022 12:15 pm - 01:15 pm INTRODUCTION: About 7% of acute pancreatitis (AP) cases are caused by hypertriglyceridemia (HTG). In such cases bowel rest, IV fluids, symptomatic therapy, and triglyceride (TG) lowering interventions are initiated. Plasmapheresis is one of the treatment options, but it has specific indications. We present a case of severe hypertriglyceridemia-induced pancreatitis that required plasmapheresis. CASE PRESENTATION: A 30 y/o man with type 2 diabetes, hyperlipidemia, multiple previous admissions for HTG-AP, presented with severe abdominal pain, nausea, and vomiting x 1 day. On admission, he was tachycardic, hypotensive, afebrile, SpO2 > 96% on RA. Labs: Glu 491 mg/dL, TG > 1000 mg/dL, Cholesterol 509 mg/dL, Lipase 987 U/L, Cr/BUN 2.4 mg/dL /20 mg/dL, VBG pH 7.25/PCO2 36.2 mmHg/PO2 19.4 mmHg/Ca 0.8/lactate 5.6;WBC 13.07 K/cm;COVID PCR positive. CXR: diffuse patchy opacities. CTAP with contrast was deferred because of AKI. He was admitted to the ICU and started on insulin drip with no improvement over 24hrs. He was still acidotic, Ca persistently low, TG still >1000, and kidney function worsened. Plasmapheresis was initiated. After one session his TG lowered to 700. He was restarted on insulin drip and in the next 24hr TG decreased to < 500 and metabolic acidosis resolved. Once AKI resolved, CT abdomen/pelvis with contrast confirmed acute pancreatitis, with focal hypodensities within the uncinate process and the proximal body, concerning infarcts as well as large phlegmon surrounding the pancreas, but no evidence of necrotizing or hemorrhagic pancreatitis. His hospital course was complicated with sepsis and DVT, which resolved with therapy. He was discharged home with TG lowering agents, Apixaban, and his previous T2DM regimen. DISCUSSION: Plasmapheresis is indicated in patients with severe HTG (>1000- 2000 mg/dl), severe HTG-AP, and when standard treatment options are inadequate. It lowers the lipid levels and removes proinflammatory markers and cytokines stopping further inflammation and damage to the pancreas and other organs faster compared to conservative therapy. Most patients need only one session which lowers TG level by 50-80%, as seen in our patient. CONCLUSIONS: Plasmapheresis should be considered in cases of HTGP with worrisome features such as lactic acidosis, hypocalcemia, worsening inflammation, and multi organ failure. Reference #1: Rajat Garg, Tarun Rustagi, "Management of Hypertriglyceridemia Induced Acute Pancreatitis", BioMed Research International, vol. 2018, Article ID 4721357, 12 pages, 2018. https://doi.org/10.1155/2018/4721357 Reference #2: Pothoulakis I, Paragomi P, Tuft M, Lahooti A, Archibugi L, Capurso G, Papachristou GI. Association of Serum Triglyceride Levels with Severity in Acute Pancreatitis: Results from an International, Multicenter Cohort Study. Digestion. 2021;102(5):809-813. doi: 10.1159/000512682. Epub 2021 Jan 21. PMID: 33477149. Reference #3: Gavva C, Sarode R, Agrawal D, Burner J. Therapeutic plasma exchange for hypertriglyceridemia induced pancreatitis: A rapid and practical approach. Transfus Apher Sci. 2016 Feb;54(1):99-102. doi: 10.1016/j.transci.2016.02.001. Epub 2016 Feb 20. PMID: 26947356. DISCLOSURES: No relevant relationships by Adam Adam No relevant relationships by Moses Bachan No relevant relationships by Chen Chao No relevant relationships by Vaishali Geedigunta No relevant relationships by Zinobia Khan No relevant relationships by Jelena Stojsavljevic

VALIDATION OF THE DIAGNOSTIC ACCURACY OF THE EPIC SEPSIS PREDICTION MODEL FOR DETECTION OF SEPSIS IN HOSPITALIZED PATIENTS

SESSION TITLE: Sepsis: Beyond 30cc/kg and Antibiotics SESSION TYPE: Rapid Fire Original Inv PRESENTED ON: 10/19/2022 11:15 am - 12:15 pm PURPOSE: The Sepsis Prediction Model (SPM) is a proprietary decision support tool created by Epic Systems. The basis of the SPM is a Predicting Sepsis Score (PSS) calculated from demographic, comorbidity, vitals, labs, medication, and procedural data. We assessed the diagnostic accuracy and timeliness of the PSS for sepsis as defined by Centers for Disease Control (CDC) Adult Sepsis Event (ASE) criteria. The performance of the PSS was compared to, Systemic Inflammatory Response Syndrome (SIRS), quick Sequential Organ Failure Assessment (qSOFA), and SOFA scores. METHODS: Retrospective review of 62,460 adults admitted to 4 Wake Forest Baptist Health System hospitals from June 1, 2019 through December 31, 2020 with PSS scores calculated every 15 minutes. A sepsis event was defined as receipt of 4 or more days of antimicrobials, blood cultures collected within 48 hours of initial antimicrobial administration, and at least one organ dysfunction. This definition of sepsis was modified to also include Covid-19 infection with organ dysfunction. Time zero was defined as time of first contact for the healthcare encounter. 30-day readmissions, facility transfers, and deaths in the Emergency Department were excluded. RESULTS: The prevalence of sepsis in the sample was 4.5%. The optimal PSS threshold based on Youden’s J statistic was a score of 8 (sensitivity 0.72, specificity 0.74, Youden’s J 0.46). SIRS (sensitivity 0.90, specificity 0.42), qSOFA (sensitivity 0.64, specificity 0.69), and SOFA (sensitivity 0.89, specificity 0.43) had a Youden’s J statistic for sepsis of 0.32, 0.33, and 0.32, respectively. At a PSS score of ≥ 8, median time to score positivity among those who reached that score (28.4% of sample) was 217 minutes (IQR 74-1477 minutes). For SIRS, qSOFA and SOFA, median time to score positivity was 54 minutes (IQR 24-456), 360 minutes (IQR 53-1593) and 107 minutes (IQR 39-474), respectively. CONCLUSIONS: Discrimination of the PSS for detection of sepsis was highest at a threshold score of 8. Overall, the PSS discriminated better than SIRS, qSOFA and SOFA. Positive SIRS and SOFA scores occurred at an earlier time-point than PSS score. The time to positivity appears to limit the tool’s best expected performance to improve time to initial antimicrobial and compliance with the 3-hour sepsis bundle. CLINICAL IMPLICATIONS: Clinical application of the Epic SPM to improve adherence with sepsis treatment goals is constrained by time to positive screen as compared to other screening tools. DISCLOSURES: No relevant relationships by Alain Bertoni No relevant relationships by Kristin Lenoir No relevant relationships by Beverly Levine No relevant relationships by Morgana Mongraw-Chaffin No relevant relationships by Adam Schertz Stock Ownership Interest relationship with Johnson & Johnson Please note: years Added 04/15/2022 by Karl Thomas, value=Ownership interest stock ownership relationship with Gilead Sciences Please note: years Added 04/15/2022 by Karl Thomas, value=Ownership Stock ownership interest relationship with Bristol-Myers Squibb Please note: years Added 04/15/2022 by Karl Thomas, value=Ownership interest Stock Ownership Interest relationship with Pfizer Please note: years Added 04/15/2022 by Karl Thomas, value=Ownership interest Stock Ownership Interest relationship with Doximity Please note: 1 year Added 04/15/2022 by Karl Thomas, value=Ownership interest No relevant relationships by Brian Wells No relevant relationships by Jack White

TUBERCULOUS PLEURAL EFFUSION MASQUERADING AS DRUG-INDUCED LUPUS

SESSION TITLE: Critical Care Presentations of TB SESSION TYPE: Rapid Fire Case Reports PRESENTED ON: 10/18/2022 12:25 pm - 01:25 pm INTRODUCTION: TNFα plays a pivotal role in inflammation and maintenance of immune response against tuberculosis. The use of TNF inhibitors (TNFi) is associated with a significant increase in the incidence of tuberculosis (TB). TNFi may cause drug-induced lupus (ATIL) presenting as constitutional symptoms, rashes, pericardial and pleural effusions with positive autoantibodies. We present a case of pleural TB masquerading as drug-induced lupus. CASE PRESENTATION: A 68y/o woman with a history of ulcerative colitis (on infliximab, mesalamine), hypertension, T2DM, CAD, complained of low-grade fever, rashes, left-sided chest pain, dyspnea, and arthralgias for two weeks. Chest pain- worse with inspiration and cough. She emigrated from India to the USA 40 years ago. Six months before infliximab therapy, Quantiferon gold was negative. Exam: faint hyperpigmentation over shins, minimal swelling of MCPs and ankles, dullness to percussion over the left chest with decreased breath sounds. Labs: CRP 101 mg/dL, Hb 10.8 iron deficient, rheumatoid factor and anti-CCP negative, ANA 1:40, dsDNA 1:640, a reminder of ENA negative, anti-histone negative, C3/C4 normal, UA bland, protein/Cr 0.4 mg/gm, negative blood cultures, SPEP and LDH normal. CXR: opacification of the left lung up to midfield. CT chest: moderate left and small right pleural effusions, enlarged mediastinal lymph nodes. COVID and Quantiferon: negative. Thoracentesis: 850 ml of exudative fluid (2 out of 3 Light's criteria), lymphocytic predominance (76% of 4148 nucleated cells), adenosine deaminase (ADA) 42 U/L, gram stain, culture, acid-fast and MTB PCR negative, cytology negative. Thoracoscopy with biopsy of the parietal pleura: necrotizing granulomatous pleuritis with acid-fast bacilli. Sensitivity: pan-sensitive M. tuberculosis. Sputum: negative for TB. She was discharged on RIPE treatment for reactivation of TB. DISCUSSION: The incidence of infliximab-induced lupus is approximately 0.19% and confirming the diagnosis is challenging. The immunogenicity of infliximab is high, 66% of patients develop positive ANA. Anti-histone antibodies are less commonly associated with ATIL as opposed to classic drug-induced lupus and dsDNA is positive in up to 90% of cases of ATIL. Renal involvement is rare. The diagnostic usefulness of ADA (over 40 U/L) in lymphocytic pleural effusions for the diagnosis of tuberculosis in an immunosuppressed individual is demonstrated here. In countries with low TB burden, such as the USA, the positive predictive value of ADA in pleural fluid declines but the negative predictive value remains high. CONCLUSIONS: Tuberculous pleuritis is not always easily diagnosed since AFB smears and sputum may remain negative. When ADA level in lymphocytic pleural fluid is not low thorough search for TB with thoracoscopy and biopsy is justified. Reference #1: Shovman O, Tamar S, Amital H, Watad A, Shoenfeld Y. Diverse patterns of anti-TNF-α-induced lupus: case series and review of the literature. Clin Rheumatol. 2018 Feb;37(2):563-568. Reference #2: Benucci, M., Gobbi, F. L., Fossi, F., Manfredi, M. & Del Rosso, A. (2005). Drug-Induced Lupus After Treatment With Infliximab in Rheumatoid Arthritis. JCR: Journal of Clinical Rheumatology, 11 (1), 47-49. Reference #3: Valdés L, San José ME, Pose A, Gude F, González-Barcala FJ, Alvarez-Dobaño JM, Sahn SA. Diagnosing tuberculous pleural effusion using clinical data and pleural fluid analysis A study of patients less than 40 years-old in an area with a high incidence of tuberculosis. Respir Med. 2010 Aug;104(8):1211-7. DISCLOSURES: No relevant relationships by Adam Adam No relevant relationships by Moses Bachan No relevant relationships by Chen Chao No relevant relationships by Zinobia Khan No relevant relationships by Milena Vukelic

Deep extracted features to support Content-Based Image Retrieval systems in the diagnosis of Covid-19 and Interstitial diseases

Purpose Coronavirus disease 2019 (Covid-19) may cause dyspnoea, whereas Interstitial Lung Diseases (ILD) may lead to the loss of breathing ability. In both cases, chest X-Ray is typically one of the initial studies to identify the diseases as they are simple and widely available scans, especially in under-development countries. However, the assessment of such images is subject to a high intraobserver variability because it depends on the reader's expertise, which may expose patients to unnecessary investigations and delay the diagnosis. Content-based Image Retrieval (CBIR) tools can bridge such a variability gap by recovering similar past cases to a given reference image from an annotated database and acting as a differential diagnosis CAD-IA system [1]. The main CBIR components are the feature extraction and the query formulation. The former represents the compared images into a space where a distance function can be applied, and the latter relies on the k-Nearest Neighbor (kNN) method to fetch the most similar cases by their distances to the query reference. In this study, we examine the quality of Covid-19 and ILD deep features extracted by a modified VGG-19 Convolutional Neural Network (CNN) [2] following the perspective of the Voronoi frontiers induced by kNN, which is at the core of the CBIR query formulation component. Methods We curated a dataset of annotated chest X-Rays from our PACS/HIS systems following a retrospective study approved by the institutional board. A set of 185 Covid-19 and 307 ILD cases from different patients was selected, being Covid-19 cases confirmed by RT-PCR tests and ILD images included after the analysis of two thoracic radiologists. We also added 381 images of ''Healthy'' lungs (without Covid-19 or ILD) to enrich the dataset. The resulting set includes 873 X-Rays (mean age 60.49 ± 15.21, and 52.58% females). We cast the DICOM images into PNG files by using the Hounsfield conversion and a 256 Gy-scale window. The files were scaled to 224 × 224 images and fed into a modified VGG-19 version we implemented [2]. Our version includes the stack of convolutional layers and five new layers after the block5-pool, namely: GlobalAveragePooling2D, BatchNormalization, a dense layer with 128 units and ReLU, a dropout layer with ratio 0.6, and a final dense layer with three neurons for classification. The Adam function was used to minimize cross-entropy, whereas batch size and epochs were set to 36 and 100, respectively. All layers start with ImageNet weights that were frozen until block4-pool so that only the remaining layers were updated. We fed the CNN with images and labels (i.e., {Covid-19, ILD, Healthy}) so that our feature extraction procedure was oriented towards those classes rather than autoencoders. The flattened outputs of the last max-pooling layer were collected as feature vectors of dimensionality d = 512. We clean and preprocess those vectors before applying the kNN-based search mechanism. First, we scaled the dimensions into the [0,1] interval. Then, we perform a reduction by using the Principal Component Analysis (PCA). The number of reduced dimensions was determined by the intrinsic dimensionality of the features, estimated by the mean (l) and standard deviation (r) of the pairwise distance distribution as the value μ2/2.σ2. Finally, the reduced vectors were also scaled into the [0,1] interval. The experiments were performed in a 3854 core 1.5 GHz GPU NVidia TitanX 12 GB RAM, and an Intel(R) Xeon(R) CPU 2.00 GHz, 96 GB RAM. The code was implemented under Tensorflow (v.2.1.0) and R (v4.1.2). Results We used two Principal Components to reduce the vectors according to the estimated intrinsic dimensionality. Figure 1 shows the Voronoi frontiers induced by kNN with a smooth separation between the three classes, which creates a search space in which CBIR searches are expected to be accurate. We quantify such behavior through a kNNbased classification on the two experimental settings (i.e., 10-folds and Holdout) by using the scaled features with and without dimensionality reductio . able 1 summarizes the results with the following findings: • The accuracy measures increased with the neighborhood (k = 1 vs. k = 5) in all experimental cases, • Covid-19 cases were more difficult to label than ILD according to F1 and RC, • The kNN hit-ratio (TP) for Covid-19 was comparable to the very first diagnosis stored into the PACS/HIS systems by readers on duty regarding the Holdout cases (readers' mean ∗ 63% vs. KNN ∼ 59%), • Searches over the reduced data were ∼ 4 9 faster, and • While dimensionality reduction was just as suitable as nonreduced data in the 10-folds evaluation, it expressively enhanced the kNN performance for the Holdout test (e.g., 0.68 vs. 0.82, k = 1 and F1). This result shows the side-effects of searching high-dimensional spaces with kNN (the ''curse of dimensionality''), which requires pre-processing the vectors or defining other query criteria to browse the data. Conclusion This study has discussed feature extraction for Covid-19 and ILD images from the perspective of kNN queries, the query formulation component within CBIR systems. Although we used cross-validation and one external batch to mitigate overfitting, a practical limitation was the size of the CNN training set. Still, our approach showed promising results in the extraction of suitable features for CBIR environments.

Relation of Anxiety and Depression with Personal Protective Equipment Associated Headache in Healthcare Workers During COVID-19 Outbreak

Objective: To determine the relationship of anxiety and depression with Personal Protective Equipment (PPE) associated headache. Background: Healthcare workers who treat COVID-19 patients were mandated to use PPE to protect themselves. Personal Protective Equipment causes some difficulties and discomfort as well as some side effects, such as headache. Personal Protective Equipment associated headache might be caused by some pathomechanisms, such as mechanic factor, hypercarbia, and stress. Anxiety and depression are thought to trigger or exacerbate PPE-associated headache. Design/Methods: This was a descriptive analytic study with cross-sectional method conducted in June 2021. We involved healthcare workers in Haji Adam Malik General Hospital, a referral centre for COVID-19 in North Sumatera, Indonesia. Personal Protective Equipment associated headache was assesed with The International Classification of Headache Disorder, 3 Edition (ICHD-3) criteria for external compression headache. Depression was screened with Patient Health Questionnaire-9 (PHQ-9). Anxiety was screened with Generalized Anxiety Disorder-7 (GAD-7) scale. Results: There were 110 respondents in this study: 64 (58,2%) were female, aged 31,94 ±6,43 years, occupation nurse 74(67,3%). Personal Protective Equipment associated headache was experienced by 64 (58,2%) respondents. It was correlated with N95 mask usage (p=0,001), goggle usage (p=0,001), female healthcare workers (p= 0,008) and nurse profession (p= 0,001). Anxiety was reported by 13 (11,8%) respondents, with 10 respondents had mild anxiety. The profession of doctor were correlated with anxiety (p=0,009). Depression was reported by 28 (25,5%) respondents, with 24 respondents had mild depression. There was no correlation between respondents characteristic with depression. Study analysis showed that anxiety and depression were not significantly associated with PPE-associated headache, with p value 0,794 dan 0,897. rd Conclusions: Anxiety and depression were not associated with PPE-associated headache in COVID-19 healthcare workers. These results suggest that another pathomechanisms have bigger influence in PPE-associated headache.

Higher Inflammatory Markers are correlated with Worse Cognitive Function in Coronavirus Disease-2019 Patients

AIM: This study aimed to determine the correlation between inflammation with cognitive function in COVID-19 patients. METHODS: We recruited COVID-19 patients using consecutive sampling methods in Adam Malik General Hospital Medan, Indonesia. The neutrophil-to-lymphocyte ratio (NLR,) C-reactive protein (CRP), D-dimer, and ferritin serum levels were measured as inflammatory markers. Cognitive function was assessed in several cognitive domains using Forward Digit Span for attention, Bacward Digit Span for working memory, and Trail Making Test parts A and B for executive function. The correlation between inflammatory markers and cognitive function was analyzed using Spearman correlation test. RESULTS: This study involved 40 COVID-19 patients consisting of 13 (32.5%) males and 27 (67.5%) females;the median age of the patients was 39.5 (19–65) years. We found that higher D-dimer and ferritin levels were significantly correlated with worse BDS scores (r = −0.369 p = 0.019 and r = −0.408 p = 0.009, respectively) and higher ferritin level was also correlated with worse FDS score (r = −0.365 p = 0.020 and). Higher D-dimer and ferritin levels were also significantly correlated with a longer time of completion of TMT-B (r = 0.363 p = 0.022 and r = 0.433 p = 0.005) and higher ferritin level was also correlated with a longer time of completion of TMT-A (r = 0.438 P=0.005). There were no significant correlations between NLR and CRP levels with cognitive function. CONCLUSION: Higher inflammatory markers are correlated with worse attention, working memory, and executive function in COVID-19 patients.

The Baseline, Clinical, and Laboratory Parameters of Breast Cancer Subjects Infected with COVID-19 in Medan, Indonesia

BACKGROUND: Cancer patients may be susceptible to COVID-19 infection due to decreased immune status. Breast cancer is the most common cancer in Indonesia, still has high admission, which increasing the risk of exposure to COVID-19. AIM: Thus, this study aimed to identify hospitalized breast cancer patients diagnosed with COVID-19 infection 1 year after the pandemic. METHODS: This is a cross-sectional study that was conducted in the Adam Malik General Hospital in Medan, Sumatera Utara, Indonesia. The enrolled subjects were those who previously histopathologically confirmed with breast cancer and having laboratory-confirmed COVID-19 infection The sources of baseline, clinical, and laboratory data were retrieved from the electronic medical records. Statistical analysis was performed using the SPSS 16.0 software (SPSS Inc., Chicago, IL, USA). RESULTS: A total of 17 female breast cancer subjects with COVID-19 infection were enrolled in this study. Mostly subjects were multiparity, highest education was junior high school, housewife, menopause, diagnosed in Stage IV, had metastasis in lung, and categorized luminal B with invasive cancer of non-special type. Most subjects showed mild clinical and radiological severity of COVID-19 infection. Low leukocyte, high neutrophil-to-lymphocyte, and high platelet-to-lymphocyte counts were significantly differed between alive and death outcome in the subjects. CONCLUSION: The baseline and clinical characteristics of female breast cancer subjects with COVID-19 infection were similar to general characteristics in the population. The parameters of leukocyte, neutrophil-to-lymphocyte, and platelet-to-lymphocyte counts could be a valuable predictive parameters of mortality outcomes.

Prevalence of Scoliosis among School going Children in Puducherry

Introduction: Scoliosis refers to the lateral curvature of the spine, which is progressive leading to morphological changes affecting quality of life, postural disability, even causing respiratory distress. Hence, screening the prevalence of scoliosis is much needed at a very early stage among school going children, thereby preventing permanent damage and spinal disorders that worsens. Aim: This study was undertaken to identify the prevalence of scoliosis among school going children in Puducherry thereby adequate adequate knowledge shall be provided to them to make them aware of the clinical condition,change their perception about the disease and motivate them to take necessary steps to overcome scoliosis at an earlier phase. Materials and Methods: A total of 1164 children (724 boys, 440 girls) aged from 6-17 years, were recruited for the study (since COVID-19 phase, most of the educational institutions were conducting only online sessions). The screening procedure included Adam's test(forward bending test) and scoliometer measurements, if >70 referred for poster anterior radiograph of the trunk, where Cobb angle is measured ,and if found to be >200 were identified to have scoliosis. The objectives were fully informed to the parents and the students and respondents were allowed to participate voluntarily. The data obtained were analyzed statistically. Results: A total of 60 students were found positive on forward bending test and scoliometer measurements >7°, among which 22 were confirmed with scoliosis on standing radiographs., 16 (73%) girls and 6 (27%) boys. Thirty-eight of which 23 (60%) girls and 15 (40%) boys had normal spine curvatures on X- ray examination (false positive). Conclusion: A total of 60 students were suffering from scoliosis, who needs further attention and the same have been informed to the concerned parents.

Relationship between Smell and Taste Impairment Function among COVID-19 Patient in General Hospital of Adam Malik Medan

BACKGROUND: Decrease of smell-taste has become a cardinal symptom of COVID-19. Some previous studies have reported that most of the COVID-19 patients complained of early smell-taste impairment. AIM: Hence, this study was aimed to investigate the relationship between smell and taste impairment against PCR test results. METHODS: This study was an observational study with a cross-sectional study among 193 patients who were diagnosed as COVID-19 that had smell-taste impairment between March and August 2020. Parameters were evaluated in this study included age, sex, ethic, occupation, smell-taste impairment, coagulation state, comorbid condition, obesity, and the result of the swab PCR test. The relationship between smell and taste impairment and the PCR test result was analyzed by Chi-square and regression logistic. RESULTS: This study showed that most COVID-19 patients were female (32.1%) aged 18−40 years old (25.9%). Meanwhile, the most comorbid condition owned by the COVID-19 patient in the General Hospital of Adam Malik Medan was diabetes, followed by hypertension, cardiovascular disease, and stroke. Moreover, this study also revealed that the smell (adjusted OR: 3.92;95% CI: 1.30−11.87) and taste (adjusted OR: 3.64;95% CI: 1.30−10.22) impairment significantly associated with COVID-19 (p < 0.05) CONCLUSION: Overall, it can be concluded that the smell-taste impairment can be used to early symptoms to predict the COVID-19 with the positive PCR test result.

Comparison of Leukocyte, Neutrophil, and Lymphocyte Levels in Child Patients that Experienced Breathlessness with COVID-19 and Non-COVID-19 (2020–2021) in Medan, Indonesia

BACKGROUND: COVID-19 is a disease caused by severe acute respiratory syndrome coronavirus-2. Various risk factors affect the morbidity and mortality of COVID-19, such as age, gender, ethnicity, comorbid conditions, and laboratory parameters. Clinical manifestations of COVID-19 in children are asymptomatic mild degrees or with acute upper respiratory tract symptoms and moderate degrees with clinical pneumonia. In the course of the COVID-19 pandemic, several biomarkers are needed that can be useful in risk stratification to predict the severity and fatality of COVID-19. Several biomarkers are used as markers of disease progression to a critical degree, such as leukocytes, lymphocytes, platelets, interleukin-6, and serum ferritin. Increased leukocytes and neutrophils and decreased lymphocytes are associated with severity and mortality in COVID-19. Laboratory findings in children with COVID-19 include decreased lymphocytes, leukopenia, and increased procalcitonin. In severe COVID-19, children often have abnormal laboratory parameters that suggest a systemic inflammatory response. AIM: This study aims to compare leukocytes, neutrophils, and lymphocytes levels in children with the complaints of shortness of breath with COVID-19 and non-COVID-19 pediatric patients. MATERIALS AND METHODS: This study is an analytic study with a cross-sectional approach to pediatric patients treated in Haji Adam Malik General Hospital Medan with the complaints of shortness of breath. Patient data were assessed based on the characteristic values of hemoglobin, leukocytes, neutrophils, lymphocytes, and COVID-19 polymerase chain reaction results and then analyzed using SPSS for Windows software. The study samples were 276 patients, with 43 confirmed positive COVID-19 patients and 233 negative patients. The source of this research data is data from the patient’s medical record for 1 year (2020–2021). RESULTS: The statistical analysis results found differences in the levels of leukocytes, neutrophils, and lymphocytes in children diagnosed with COVID-19 and non-COVID-19 (p < 0.05). However, there was no difference in hemoglobin levels in children who were congested with COVID-19 and non-COVID-19 (p > 0.05). CONCLUSION: These results are consistent with several cases and previous studies where the levels of leukocytes, neutrophils, and lymphocytes can be used as a detection tool and a diagnostic and prognostic tool in COVID-19 patients.

Introduction of CDISC therapeutic user guide for COVID-19 and its related standards

【】 The coronavirus disease 2019 (COVID-19) has been evaluated as a pandemic by the World Health Organization (WHO). Although several vaccines with an efficacy of more than 50% have been developed, in terms of treatment, remdesivir remains the antiviral drug approved by the U.S. Food and Drug Administration (FDA) for the treatment of COVID-19, while none of the other treatments has been recommended by FDA due to insufficient clinical data. A number of clinical trials have been registered to study therapeutic drugs or vaccines for COVID-19. To promote the collection, tabulation, analysis of COVID-19 clinical research data, improve the efficiency of clinical research, and facilitate the integration, sharing and secondary analysis of multiple similar research data, CDISC has developed a therapeutic area user guide for COVID-19 and resources for public health researchers. The resources included two documents, one is a SDTM annotated CRF based on ISARIC and WHO core COVID-19 case report form, and the other is SDTM and CDASH mapping spreadsheet. Moreover, CDISC has developed the guidance for ongoing studies disrupted by COVID-19 pandemic and interim ADaM guidance for ongoing studies disrupted by COVID-19 pandemic to help disrupted clinical trials to collect, store, and analyze relevant data. This paper introduced the structure and content of the guide and its related standards, with a view to promoting its application in COVID-19 clinical trials and in ongoing studies disrupted by COVID-19 pandemic.