Erythrocyte miRNA-92a-3p interactions with PfEMP1 as determinants of clinical malaria.

Based on the recently added high throughput analysis data on small noncoding RNAs in modulating disease pathophysiology of malaria, we performed an integrative computational analysis for exploring the role of human-host erythrocytic microRNAs (miRNAs) and their influence on parasite survival and host homeostasis. An in silico analysis was performed on transcriptomic datasets accessed from PlasmoDB and Gene Expression Omnibus (GEO) repositories analyzed using miRanda, miRTarBase, mirDIP, and miRDB to identify the candidate miRNAs that were further subjected to network analysis using MCODE and DAVID. This was followed by immune infiltration analysis and screening for RNA degradation mechanisms. Seven erythrocytic miRNAs, miR-451a, miR-92a-3p, miR-16-5p, miR-142-3p, miR-15b-5p, miR-19b-3p, and miR-223-3p showed favourable interactions with parasite genes expressed during blood stage infection. The miR-92a-3p that targeted the virulence gene PfEMP1 showed drastic reduction during infection. Performing pathway analysis for the human-host gene targets for the miRNA identified TOB1, TOB2, CNOT4, and XRN1 genes that are associated to RNA degradation processes, with the exoribonuclease XRN1, highly enriched in the malarial samples. On evaluating the role of exoribonucleases in miRNA degradation further, the pattern of Plasmodium falciparum_XRN1 showed increased levels during infection thus suggesting a defensive role for parasite survival. This study identifies miR-92a-3p, a member of C13orf25/ miR-17-92 cluster, as a novel miRNA inhibitor of the crucial parasite genes responsible for symptomatic malaria. Evidence for a plausible link to chromosome 13q31.3 loci controlling the epigenetic disease regulation is also suggested.

Effect of targeted intervention on C-terminal agrin fragment and its association with the components of sarcopenia: a scoping review.

BACKGROUND: C-terminal Agrin Fragment (CAF) has emerged as a potent biomarker for identifying sarcopenia. However, the effect of interventions on CAF concentration and the association of CAF with sarcopenia components are unclear. OBJECTIVE: To review the association between CAF concentration and muscle mass, muscle strength, and physical performance among individuals with primary and secondary sarcopenia and to synthesize the effect of interventions on the change in the level of CAF concentration. METHODS: A systematic literature search was conducted in six electronic databases, and studies were included if they met the selection criteria decided a priori. The data extraction sheet was prepared, validated, and extracted relevant data. RESULTS: A total of 5,158 records were found, of which 16 were included. Among studies conducted on individuals with primary sarcopenia, muscle mass was significantly associated with CAF levels, followed by hand grip strength (HGS) and physical performance, with more consistent findings in males. While in secondary sarcopenics, the strongest association was found for HGS and CAF levels, followed by physical performance and muscle mass. CAF concentration was reduced in trials that used functional, dual task, and power training, whereas resistance training and physical activity raised CAF levels. Hormonal therapy did not affect serum CAF concentration. CONCLUSION(S): The association between CAF and sarcopenic assessment parameters varies in primary and secondary sarcopenics. The findings would help practitioners and researchers choose the best training mode/parameters/exercises to reduce CAF levels and, eventually, manage sarcopenia.

Psychometric performance of the Kannada version of sarcopenia quality of life questionnaire (SarQoL®).

BACKGROUND: The Sarcopenia Quality of Life (SarQoL®) is a patient reported quality-of-life questionnaire specific to sarcopenia. In the Indian context, its availability is limited to Hindi, Marathi and Bengali vernacular languages. AIMS: This study aimed to translate, cross-culturally adapt the SarQoL® questionnaire into Kannada and investigate its psychometric properties. METHODS: The SarQoL®-English version was translated into Kannada with the developer's permission and in accordance with their requirements. To validate the discriminative power, internal consistency and floor and ceiling effect of the SarQoL®-Kannada questionnaire were assessed in the first step. In the second step, the construct validity and the test-retest reliability of the SarQoL®-Kannada was determined. RESULT: There was no difficulty in the translation process. A total of n = 114 participants (sarcopenic participants n = 45 and n = 69 non-sarcopenic participants) were included. The good discriminative power of the SarQoL®-Kannada questionnaire {quality of life for sarcopenic subjects [56.43 ± 11.32] vs. non-sarcopenic ones [79.38 ± 8.16], p < 0.001}. High internal consistency (Cronbach's alpha coefficient was 0.904) and no ceiling/ floor effect were reflected. Excellent test-retest reliability (intraclass correlation coefficient was 0.97, 95% CI 0.92-0.98) were found. A good convergent and divergent validity with similar and different domains of WHOQOL-BREF was observed, while EQ-5D-3L had good convergent and weak divergent validity. CONCLUSION: The SarQoL®-Kannada questionnaire is valid, consistent and reliable for the measurement of quality of life of sarcopenic participants. SarQoL®-Kannada questionnaire is now available to be used in clinical practice and as a treatment outcome indicator in research.

Ethnic disparities attributed to the manifestation in and response to type 2 diabetes: insights from metabolomics.

Type 2 diabetes (T2D) associated health disparities among different ethnicities have long been known. Ethnic variations also exist in T2D related comorbidities including insulin resistance, vascular complications and drug response. Genetic heterogeneity, dietary patterns, nutrient metabolism and gut microbiome composition attribute to ethnic disparities in both manifestation and progression of T2D. These factors differentially regulate the rate of metabolism and metabolic health. Metabolomics studies have indicated significant differences in carbohydrate, lipid and amino acid metabolism among ethnicities. Interestingly, genetic variations regulating lipid and amino acid metabolism might also contribute to inter-ethnic differences in T2D. Comprehensive and comparative metabolomics analysis between ethnicities might help to design personalized dietary regimen and newer therapeutic strategies. In the present review, we explore population based metabolomics data to identify inter-ethnic differences in metabolites and discuss how (a) genetic variations, (b) dietary patterns and (c) microbiome composition may attribute for such differences in T2D.

Perceptions of family-centred care among caregivers of children with cerebral palsy in South India: An exploratory study.

BACKGROUND: Parental views and expectations about family-centred care (FCC) need to be understood for its successful implementation. Knowledge of caregiver's perceptions and needs, within their social and cultural context, forms the basis for effective health care partnerships with families. The purpose of this study was to explore perceptions about FCC among caregivers of children with cerebral palsy (CP) in South India. METHOD: Fourteen caregivers of children with CP (aged 4-12 years) living in rural areas of the coastal region of Karnataka, India, participated in this qualitative study. Face-to-face semi-structured interviews were recorded and transcribed for analysis using the framework and hybrid approaches. Thematic mapping of the categories and themes was done to explore relationships about perceptions of FCC. RESULTS: Caregiver's life emerged inductively as a new theme highlighting caregiver's own physical and mental health, family roles and well-being, integrating the onus of care with household responsibilities, limited participation in personal activities and social isolation. The qualitative findings revealed the ubiquity of respectful and trusting relationships with health professionals while expressing paucity of coordinated comprehensive care, sporadic partnerships and opportunities for shared decision-making; desire for receiving specific information related to child's progress and prognosis; and general information on community resources and the need of empowerment and support groups. CONCLUSION: Our study has practical implications for the implementation of FCC within the South Indian context, by recognizing unique caregiver needs and expectations in sync with cultural perspectives towards childhood disability such as societal stigma, values and traditional beliefs; attitudes towards medical professionals; and life stressors and gender responsibilities.

Transmission of SARS-CoV-2 Infection by Children: A Study of Contacts of Index Paediatric Cases in India.

The susceptibility of children to coronavirus disease-19 (COVID-19) and transmission of COVID-19 from children to others is a relatively unexplored area. The aim of this study was to understand the transmission dynamics of Severe Acute Respiratory Syndrome Coronavirus 2 in children. This was a retrospective observational study where a total of 19 paediatric index cases (including a set of twins) with COVID-19 and 42 primary contacts (adults-36, paediatric-6) from the immediate family members were included. All the index cases and four of the five positive contacts were asymptomatic. Despite adults staying with positive children in the same vehicle, same room in the quarantine centre and the same ward, only four of the parents became positive.

Neutrophil-Lymphocyte Ratio in Patients with COVID-19 as a Simple Tool to Predict Requirement of Admission to a Critical Care Unit.

INTRODUCTION: Coronavirus disease-2019 (COVID-19) pandemic has overloaded the healthcare system beyond its functional capacity. Late referral to higher levels of care may be one of the factors associated with higher mortality. Therefore, we aimed to find simple demographic and laboratory parameters which predict the requirement of admission to a critical care unit. MATERIALS AND METHODS: A case-control study was undertaken in adult age population >18 years, admitted in a dedicated COVID hospital in South India. A total of 50 patients with severe disease (cases) were compared with 143 mild or asymptomatic cases (controls). Those demographic and laboratory parameters that were found to be significant on univariate analysis were used for multiple logistic regression analysis. RESULTS: Univariate analysis of demographic and laboratory data showed higher age, male sex, presence of diabetes mellitus, higher values of C-reactive protein, ferritin, D-dimer, neutrophil-lymphocyte ratio (NLR), and lactate dehydrogenase to be significantly associated with cases. Multivariate logistic regression analysis of these significant variables showed NLR and ferritin to be the independent predictors of the requirement of admission to a critical care unit. The receiver-operating characteristic curve showed an NLR value of 5.2 and a ferritin value of 462 µg/L that were able to predict the requirement of admission in critical care units. CONCLUSION: High ferritin and NLR were independent predictors of the requirement of admission in critical care units. NLR is a simple tool that can be used in resource-limited settings for triage and early referral to higher levels of care. HOW TO CITE THIS ARTICLE: Maddani SS, Gupta N, Umakanth S, Joylin S, Saravu K. Neutrophil-Lymphocyte Ratio in Patients with COVID-19 as a Simple Tool to Predict Requirement of Admission to a Critical Care Unit. Indian J Crit Care Med 2021;25(5):535-539.

CpG-SNP site methylation regulates allele-specific expression of MTHFD1 gene in type 2 diabetes.

The interaction of genetic and epigenetic mechanisms is one of the underlying causes of phenotypic variability in complex diseases such as type 2 diabetes (T2D). To explore the influence of genetic and epigenetic changes in T2D, we examined the effect of methylation of CpG-SNP sites on allele-specific expression (ASE) in one-carbon metabolism pathway genes in T2D. Case-control study was conducted on 860 individuals (430 T2D and 430 controls). CpG-SNPs shortlisted through in silico analysis were genotyped using tetra ARMS PCR and validated using Sanger DNA sequencing. Global DNA methylation was carried out using RP-HPLC. Promoter DNA methylation and CpG site-specific methylation were carried out using bisulfite sequencing. mRNA expression and ASE were examined by SYBR green and TaqMan assay, respectively. Four exonic CpG-SNPs of MTHFD1, MTRR, and GGH genes were identified in folate pathway genes. Among these, MTHFD1 rs2236225 showed significant association with T2D independent of obesity, displayed ASE, and correlated with CpG-SNP site-specific methylation when compared with controls. Our results demonstrate that SNP rs2236225 in the CpG site of MTHFD1, which regulates allele-specific gene expression in PBMCs is methylation dependent and may perturb one-carbon metabolism pathway in T2D subjects.

Using an integrated teaching approach to facilitate student achievement of the learning outcomes in a preclinical medical curriculum in India.

Today most education institutions around the world have adopted the philosophy of outcome-based education. The emphasis in outcome-based education is achievement of outcomes; hence the curriculum should be designed in a way that it includes the components targeted specifically at achieving these outcomes. A discipline-based approach results in fragmentation of learning and lack of clinical applicability. Integrated teaching could be a solution to achieve required outcomes in a holistic way. Hence, the aim of this study was to develop, implement, and evaluate an integrated teaching module. Temporal coordination of the basic sciences, along with correlation of learned topics to clinical settings, was done in the first year of the undergraduate medical program. The module was evaluated by obtaining qualitative and quantitative feedback from students. Student assessment was conducted with a test that had case vignettes and multiple-choice questions. In addition, students' change in learning approaches and self-directed learning readiness were collected. Students' perception regarding the educational environment was also obtained. Analysis of the data showed positive feedback from the students regarding the integrated teaching. Students' average score in the test was 86%. There was a significant increase in the scores for the deep approach and self-directed learning readiness in the posttest compared with the pretest. Moreover, students were found to be satisfied with the educational environment. Evaluation of integrated teaching revealed that it was well accepted by the students. Moreover, it facilitated the achievement of the students' outcomes.

A pilot randomized controlled trial to compare the effectiveness of two 14-day primaquine regimens for the radical cure of vivax malaria in South India.

BACKGROUND: Radical cure of Plasmodium vivax malaria requires treatment with a blood schizonticide and a hypnozoitocide (primaquine) to eradicate the dormant liver stages. There has been uncertainty about the operational effectiveness and optimum dosing of the currently recommended 14-day primaquine (PQ) course. METHODS: A two centre, randomized, open-label, two arm study was conducted in South India. Patients were randomized to receive either high dose (0.5 mg base/kg body weight) or conventional dose (0.25 mg/kg) PQ for 14 days. Plasma concentrations of PQ and carboxyprimaquine (CPQ) on the 7th day of treatment were measured by reverse phase high performance liquid chromatography. Study subjects were followed up for 6 months. Recurrent infections were genotyped using capillary fragment length polymorphism of two PCR-amplified microsatellite markers (MS07 and MS 10). RESULTS: Fifty patients were enrolled. Baseline characteristics and laboratory features did not differ significantly between the groups. Mean age of the study population was 42 ± 16.0 years. Recurrences 80-105 days later occurred in 4 (8%) patients, two in each the groups. All recurrences had the same microsatellite genotype as that causing the index infection suggesting all were relapses. One relapse was associated with low CPQ concentrations suggesting poor adherence. CONCLUSIONS: This small pilot trial supports the effectiveness of the currently recommended lower dose (0.25 mg/kg/day) 14 day PQ regimen for the radical cure of vivax malaria in South India. Trial registration Clinical Trials Registry-India, CTRI/2017/03/007999. Registered 3 March 2017, http://ctri.nic.in/Clinicaltrials/regtrial.php?modid=1&compid=19&EncHid=82755.86366 .

Effect of a Customized Physical Activity Promotion Program on Visceral Fat and Glycemic Parameters in Individuals With Prediabetes: A Randomized Controlled Trial.

BACKGROUND: Physical activity of any amount results in substantial health benefits. However, public awareness of physical activity benefits in chronic diseases is inadequate in India. Prediabetes is a significant health issue on a global scale. Visceral fat (VF) is considered as an early predictor of prediabetes. Ethnicity and race have a substantial impact on VF. Hence, this study intended to evaluate the effect of a customized physical activity promotion program on VF and glycemic parameters in individuals with prediabetes. METHODS: In the current, parallel group randomized controlled trial, a total of 158 participants were recruited: 79 in intervention and 79 in control group. The study included the prediabetes individuals based on American Diabetes Association criteria. Participants from the intervention group received the customized physical activity promotion program for 24 weeks. The primary outcome measures of the study were VF level and glycemic parameters that included fasting blood sugar and glycosylated hemoglobin. Two-way mixed analysis of variance was used to study the mean difference of an outcome between 2 groups over time. RESULTS: The study found a statistically significant interaction between the intervention and times on VF level, F1,136 = 23.564, fasting blood sugar levels, F1,136 = 8.762, and glycosylated hemoglobin levels, F1,136 = 64.582 at the end of 24 weeks (P < .05). CONCLUSIONS: This study concluded that a customized physical activity promotion program was effective in reducing VF in individuals with prediabetes as compared with controls. It improved glycemic control by reducing fasting blood sugar and glycosylated hemoglobin levels.

Four-Step Co-Designing of the Reablement Strategies Targeting Sarcopenia (ReStart-S): An Exercise-Based Multicomponent Program for Older Adults Residing in Long-Term Care Settings.

Background: The prevalence of sarcopenia is concerningly high in long-term care settings (LTCS); yet, no exercise programs specifically targeting older adults living in residential care are available. Objective: The goal of the present study was to co-design and validate a program named Reablement Strategies targeting Sarcopenia (ReStart-S) for older long-term care residents. Design: Cross-sectional study with an exploratory phase. Settings: LTCS in Udupi, Karnataka, India. Participants: Sarcopenic older adults diagnosed using Asian Working Group for Sarcopenia 2019 criteria. Material and Methods: The program was designed using a four-step intervention mapping technique involving systematic progression after completing each step. The steps included 1) identifying the appropriate exercise-based intervention for sarcopenia, 2) determining objectives and expected outcomes, 3) seeking expert views through a Delphi consensus approach, and 4) assessing the feasibility of ReStart-S program among older adults living in LTCS. Results: A comprehensive literature review appraised existing exercise programs for managing sarcopenia. A workshop held with six older adults and one caretaker, decided on morning exercise sessions, recommended 2-7 days/week. The results of the review and workshop were compiled for the Delphi process that had seven experts from 5 countries, achieving a 71% response rate after four rounds. In the last step, a pilot study on eight LTCS residents, two males and six females with a mean age of 78.3 ± 8.3 years, was conducted and the program was found to be feasible. Conclusion: The ReStart-S program for managing sarcopenia among older adults residing in LTCS incorporates evidence from the literature and the engagement of older adults, caregivers, and experts, making it a contextually appropriate intervention. Our study also provides researchers and healthcare professionals insight into co-designing an intervention program for vulnerable older adults. Finally, the program evaluation indicates that a full-scale trial testing the efficacy of the ReStart-S program is feasible.

Explainable artificial intelligence approaches for COVID-19 prognosis prediction using clinical markers.

The COVID-19 influenza emerged and proved to be fatal, causing millions of deaths worldwide. Vaccines were eventually discovered, effectively preventing the severe symptoms caused by the disease. However, some of the population (elderly and patients with comorbidities) are still vulnerable to severe symptoms such as breathlessness and chest pain. Identifying these patients in advance is imperative to prevent a bad prognosis. Hence, machine learning and deep learning algorithms have been used for early COVID-19 severity prediction using clinical and laboratory markers. The COVID-19 data was collected from two Manipal hospitals after obtaining ethical clearance. Multiple nature-inspired feature selection algorithms are used to choose the most crucial markers. A maximum testing accuracy of 95% was achieved by the classifiers. The predictions obtained by the classifiers have been demystified using five explainable artificial intelligence techniques (XAI). According to XAI, the most important markers are c-reactive protein, basophils, lymphocytes, albumin, D-Dimer and neutrophils. The models could be deployed in various healthcare facilities to predict COVID-19 severity in advance so that appropriate treatments could be provided to mitigate a severe prognosis. The computer aided diagnostic method can also aid the healthcare professionals and ease the burden on already suffering healthcare infrastructure.

DNA methyltransferase isoforms regulate endothelial cell exosome proteome composition.

Extrinsic and intrinsic pathological stimuli in vascular disorders induce DNA methylation based epigenetic reprogramming in endothelial cells, which leads to perturbed gene expression and subsequently results in endothelial dysfunction (ED). ED is also characterized by release of exosomes with altered proteome leading to paracrine interactions in vasculature and subsequently contributing to manifestation, progression and severity of vascular complications. However, epigenetic regulation of exosome proteome is not known. Hence, our present study aimed to understand influence of DNA methylation on exosome proteome composition and their influence on endothelial cell (EC) function. DNMT isoforms (DNMT1, DNMT3A, and DNMT3B) were overexpressed using lentivirus in ECs. Exosomes were isolated and characterized from ECs overexpressing DNMT isoforms and C57BL/6 mice plasma treated with 5-aza-2'-deoxycytidine. 3D spheroid assay was performed to understand the influence of exosomes derived from cells overexpressing DNMTs on EC functions. Further, the exosomes were subjected to TMT labelled proteomics analysis followed by validation. 3D spheroid assay showed increase in the pro-angiogenic activity in response to exosomes derived from DNMT overexpressing cells which was impeded by inclusion of 5-aza-2'-deoxycytidine. Our results showed that exosome proteome and PTMs were significantly modulated and were associated with dysregulation of vascular homeostasis, metabolism, inflammation and endothelial cell functions. In vitro and in vivo validation showed elevated DNMT1 and TGF-ß1 exosome proteins due to DNMT1 and DNMT3A overexpression, but not DNMT3B which was mitigated by 5-aza-2'-deoxycytidine indicating epigenetic regulation. Further, exosomes induced ED as evidenced by reduced expression of phospho-eNOSser1177. Our study unveils epigenetically regulated exosome proteins, aiding management of vascular complications.

Barriers and Enablers for Physical Activity Engagement Among Individuals From India With Type 2 Diabetes Mellitus: A Mixed-Method Study.

BACKGROUND: Type 2 diabetes mellitus (T2DM) is a complex, chronic condition that can cause multiple complications due to poor glycemic control. Self-management plays a crucial role in the management of T2DM. Lifestyle modifications, including physical activity (PA), are fundamental for self-management. This study explored the knowledge, perception, practice, enablers, and barriers of PA among individuals with T2DM. METHODS: A mixed-method study was conducted among individuals with T2DM in Udupi taluk, India. A cross-sectional survey (n = 467) followed by an in-depth interview (n = 35) was performed. The data were analyzed using descriptive statistics and thematic analysis, respectively. RESULTS: About half (48.8%) of the participants engaged in PA of which 28.3% had an adequate score in the practice of PA. Walking was the most preferred mode. Self-realization, Comprehension, perception, and source of information, PA training, Current PA practices, enablers and barriers for PA were 6 themes derived under knowledge, perception, and practice of PA. CONCLUSION: Despite knowing the importance of PA, compliance with PA was poor. The personal/internal, societal, and external factors constituted the trinity of barriers and enablers in compliance with PA. Behavioral changes, societal changes, policy initiatives, and PA training in health care settings may enhance PA practice among individuals with T2DM.

Photobiomodulation of surgical wound dehiscence in a diabetic individual by low-level laser therapy following median sternotomy.

In this single case study, we attempt to outline the possible effect of low-level laser therapy (LLLT) on delayed wound healing and pain in chronic dehiscent sternotomy of a diabetic individual. The methods that were employed to evaluate changes pre and post irradiation were wound photography, wound area measurement, pressure ulcer scale of healing (PUSH), and visual analogue scale (VAS) for pain. After irradiation, proliferation of healthy granulation tissue was observed with decrease in scores of PUSH for sternal dehiscence and VAS for bilateral shoulders and sternal dehiscence. We found that LLLT irradiation could be a novel method of treatment for chronic sternal dehiscence following coronary artery bypass grafting, as it augments wound healing with an early closure of the wound deficit. Hence, this might be translated into an early functional rehabilitation and decreased pain perception of an individual following surgical complication.

Artificial intelligence for diagnosis of mild-moderate COVID-19 using haematological markers.

OBJECTIVE: The persistent spread of SARS-CoV-2 makes diagnosis challenging because COVID-19 symptoms are hard to differentiate from those of other respiratory illnesses. The reverse transcription-polymerase chain reaction test is the current golden standard for diagnosing various respiratory diseases, including COVID-19. However, this standard diagnostic method is prone to erroneous and false negative results (10% -15%). Therefore, finding an alternative technique to validate the RT-PCR test is paramount. Artificial intelligence (AI) and machine learning (ML) applications are extensively used in medical research. Hence, this study focused on developing a decision support system using AI to diagnose mild-moderate COVID-19 from other similar diseases using demographic and clinical markers. Severe COVID-19 cases were not considered in this study since fatality rates have dropped considerably after introducing COVID-19 vaccines. METHODS: A custom stacked ensemble model consisting of various heterogeneous algorithms has been utilized for prediction. Four deep learning algorithms have also been tested and compared, such as one-dimensional convolutional neural networks, long short-term memory networks, deep neural networks and Residual Multi-Layer Perceptron. Five explainers, namely, Shapley Additive Values, Eli5, QLattice, Anchor and Local Interpretable Model-agnostic Explanations, have been utilized to interpret the predictions made by the classifiers. RESULTS: After using Pearson's correlation and particle swarm optimization feature selection, the final stack obtained a maximum accuracy of 89%. The most important markers which were useful in COVID-19 diagnosis are Eosinophil, Albumin, T. Bilirubin, ALP, ALT, AST, HbA1c and TWBC. CONCLUSION: The promising results suggest using this decision support system to diagnose COVID-19 from other similar respiratory illnesses.

Disseminated tuberculosis masquerading as depression.

We report the case of a 23-year-old man without any significant premorbid conditions initially presenting to a psychiatrist with suspected depression but later referred to our hospital owing to the possibility of systemic disease and subsequently diagnosed as having disseminated tuberculosis.

A Decision Support System for Diagnosis of COVID-19 from Non-COVID-19 Influenza-like Illness Using Explainable Artificial Intelligence.

The coronavirus pandemic emerged in early 2020 and turned out to be deadly, killing a vast number of people all around the world. Fortunately, vaccines have been discovered, and they seem effectual in controlling the severe prognosis induced by the virus. The reverse transcription-polymerase chain reaction (RT-PCR) test is the current golden standard for diagnosing different infectious diseases, including COVID-19; however, it is not always accurate. Therefore, it is extremely crucial to find an alternative diagnosis method which can support the results of the standard RT-PCR test. Hence, a decision support system has been proposed in this study that uses machine learning and deep learning techniques to predict the COVID-19 diagnosis of a patient using clinical, demographic and blood markers. The patient data used in this research were collected from two Manipal hospitals in India and a custom-made, stacked, multi-level ensemble classifier has been used to predict the COVID-19 diagnosis. Deep learning techniques such as deep neural networks (DNN) and one-dimensional convolutional networks (1D-CNN) have also been utilized. Further, explainable artificial techniques (XAI) such as Shapley additive values (SHAP), ELI5, local interpretable model explainer (LIME), and QLattice have been used to make the models more precise and understandable. Among all of the algorithms, the multi-level stacked model obtained an excellent accuracy of 96%. The precision, recall, f1-score and AUC obtained were 94%, 95%, 94% and 98% respectively. The models can be used as a decision support system for the initial screening of coronavirus patients and can also help ease the existing burden on medical infrastructure.

COVID-19 diagnosis using clinical markers and multiple explainable artificial intelligence approaches: A case study from Ecuador.

The COVID-19 pandemic erupted at the beginning of 2020 and proved fatal, causing many casualties worldwide. Immediate and precise screening of affected patients is critical for disease control. COVID-19 is often confused with various other respiratory disorders since the symptoms are similar. As of today, the reverse transcription-polymerase chain reaction (RT-PCR) test is utilized for diagnosing COVID-19. However, this approach is sometimes prone to producing erroneous and false negative results. Hence, finding a reliable diagnostic method that can validate the RT-PCR test results is crucial. Artificial intelligence (AI) and machine learning (ML) applications in COVID-19 diagnosis has proven to be beneficial. Hence, clinical markers have been utilized for COVID-19 diagnosis with the help of several classifiers in this study. Further, five different explainable artificial intelligence techniques have been utilized to interpret the predictions. Among all the algorithms, the k-nearest neighbor obtained the best performance with an accuracy, precision, recall and f1-score of 84%, 85%, 84% and 84%. According to this study, the combination of clinical markers such as eosinophils, lymphocytes, red blood cells and leukocytes was significant in differentiating COVID-19. The classifiers can be utilized synchronously with the standard RT-PCR procedure making diagnosis more reliable and efficient.