Validating polyp and instrument segmentation methods in colonoscopy through Medico 2020 and MedAI 2021 Challenges.

Automatic analysis of colonoscopy images has been an active field of research motivated by the importance of early detection of precancerous polyps. However, detecting polyps during the live examination can be challenging due to various factors such as variation of skills and experience among the endoscopists, lack of attentiveness, and fatigue leading to a high polyp miss-rate. Therefore, there is a need for an automated system that can flag missed polyps during the examination and improve patient care. Deep learning has emerged as a promising solution to this challenge as it can assist endoscopists in detecting and classifying overlooked polyps and abnormalities in real time, improving the accuracy of diagnosis and enhancing treatment. In addition to the algorithm's accuracy, transparency and interpretability are crucial to explaining the whys and hows of the algorithm's prediction. Further, conclusions based on incorrect decisions may be fatal, especially in medicine. Despite these pitfalls, most algorithms are developed in private data, closed source, or proprietary software, and methods lack reproducibility. Therefore, to promote the development of efficient and transparent methods, we have organized the "Medico automatic polyp segmentation (Medico 2020)" and "MedAI: Transparency in Medical Image Segmentation (MedAI 2021)" competitions. The Medico 2020 challenge received submissions from 17 teams, while the MedAI 2021 challenge also gathered submissions from another 17 distinct teams in the following year. We present a comprehensive summary and analyze each contribution, highlight the strength of the best-performing methods, and discuss the possibility of clinical translations of such methods into the clinic. Our analysis revealed that the participants improved dice coefficient metrics from 0.8607 in 2020 to 0.8993 in 2021 despite adding diverse and challenging frames (containing irregular, smaller, sessile, or flat polyps), which are frequently missed during a routine clinical examination. For the instrument segmentation task, the best team obtained a mean Intersection over union metric of 0.9364. For the transparency task, a multi-disciplinary team, including expert gastroenterologists, accessed each submission and evaluated the team based on open-source practices, failure case analysis, ablation studies, usability and understandability of evaluations to gain a deeper understanding of the models' credibility for clinical deployment. The best team obtained a final transparency score of 21 out of 25. Through the comprehensive analysis of the challenge, we not only highlight the advancements in polyp and surgical instrument segmentation but also encourage subjective evaluation for building more transparent and understandable AI-based colonoscopy systems. Moreover, we discuss the need for multi-center and out-of-distribution testing to address the current limitations of the methods to reduce the cancer burden and improve patient care.

Feasibility of deep learning-based tumor segmentation for target delineation and response assessment in grade-4 glioma using multi-parametric MRI.

Background: Tumor burden assessment is essential for radiation therapy (RT), treatment response evaluation, and clinical decision-making. However, manual tumor delineation remains laborious and challenging due to radiological complexity. The objective of this study was to investigate the feasibility of the HD-GLIO tool, an ensemble of pre-trained deep learning models based on the nnUNet-algorithm, for tumor segmentation, response prediction, and its potential for clinical deployment. Methods: We analyzed the predicted contrast-enhanced (CE) and non-enhancing (NE) HD-GLIO output in 49 multi-parametric MRI examinations from 23 grade-4 glioma patients. The volumes were retrospectively compared to corresponding manual delineations by 2 independent operators, before prospectively testing the feasibility of clinical deployment of HD-GLIO-output to a RT setting. Results: For CE, median Dice scores were 0.81 (95% CI 0.71-0.83) and 0.82 (95% CI 0.74-0.84) for operator-1 and operator-2, respectively. For NE, median Dice scores were 0.65 (95% CI 0.56-0,69) and 0.63 (95% CI 0.57-0.67), respectively. Comparing volume sizes, we found excellent intra-class correlation coefficients of 0.90 (P < .001) and 0.95 (P < .001), for CE, respectively, and 0.97 (P < .001) and 0.90 (P < .001), for NE, respectively. Moreover, there was a strong correlation between response assessment in Neuro-Oncology volumes and HD-GLIO-volumes (P < .001, Spearman's R2 = 0.83). Longitudinal growth relations between CE- and NE-volumes distinguished patients by clinical response: Pearson correlations of CE- and NE-volumes were 0.55 (P = .04) for responders, 0.91 (P > .01) for non-responders, and 0.80 (P = .05) for intermediate/mixed responders. Conclusions: HD-GLIO was feasible for RT target delineation and MRI tumor volume assessment. CE/NE tumor-compartment growth correlation showed potential to predict clinical response to treatment.

The polymorphic landscape analysis of GATA1 exons uncovered the genetic variants associated with higher thrombocytopenia in dengue patients.

The current study elucidated an association between gene variants and thrombocytopenia through the investigation of the exonic polymorphic landscape of hematopoietic transcription factor-GATA1 gene in dengue patients. A total of 115 unrelated dengue patients with dengue fever (DF) (N = 91) and dengue hemorrhagic fever (DHF) (N = 24) were included in the study. All dengue patients were confirmed through detection of NS1 antigen, IgM, and IgG antibodies against the dengue virus. Polymerase chain reaction using specific primers amplified the exonic regions of GATA1 while Sanger sequencing and chromatogram analyses facilitated the identification of variants. Variants G>A (at chX: 48792009) and C>A (at chX: 4879118) had higher frequency out of 13 variants identified (3 annotated and 10 newly recognized). Patients carrying either nonsynonymous or synonymous variants had significantly lower mean values of platelets compared to those harboring the reference nucleotides (NC_000023.11). Further analyses revealed that the change in amino acid residue leads to the altered three-dimensional structure followed by interaction with neighboring residues. Increased stability of the protein due to substitution of serine by asparagine (S129N at chX: 48792009) may cause increased rigidity followed by reduced structural flexibility which may ultimately disturb the dimerization (an important prerequisite for GATA1 to perform its biological activity) process of the GATA1 protein. This, in turn, may affect the function of GATA1 followed by impaired production of mature platelets which may be reflected by the lower platelet counts in individuals with such variation. In summary, we have identified new variants within the GATA1 gene which were found to be clinically relevant to the outcome of dengue patients and thus, have the potential as candidate biomarkers for the determination of severity and prognosis of thrombocytopenia caused by dengue virus. However, further validation of this study in a large number of dengue patients is warranted. Trial Registration: number SLCTR/2019/037.

Genome Sequences of 25 SARS-CoV-2 Sublineage B.1.1.529 Omicron Strains in Bangladesh.

We report the coding-complete genome sequences of 25 severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) sublineage B.1.1.529 Omicron strains obtained from Bangladeshi individuals in samples collected between December 2021 and January 2022. Genomic data were generated by Nanopore sequencing using the amplicon sequencing approach developed by the ARTIC Network.

Investigation of the efficacy and safety of eltrombopag to correct thrombocytopenia in moderate to severe dengue patients - a phase II randomized controlled clinical trial.

BACKGROUND: The dengue-infected patients with or without hemorrhagic manifestations, typically exhibit moderate to severe thrombocytopenia. A thrombopoietin receptor agonist - eltrombopag has been efficacious in correcting thrombocytopenia in patients with various pathological conditions including immune thrombocytopenia, chronic liver disease, and severe aplastic anemia. This study investigated the efficacy and safety of eltrombopag to correct dengue-mediated thrombocytopenia. METHODS: In this open-label, randomized controlled phase-II trial, patients with dengue fever (DF) and dengue hemorrhagic fever (DHF) having platelet (PLT) count lower than 100 × 109/L without comorbidity, pregnancy, and liver abnormalities were enrolled in Dhaka Medical College Hospital, Better Life Hospital and AMZ hospital, Dhaka, Bangladesh. Between October 10, 2019, and December 30, 2019, 123 DF and DHF patients were assessed for eligibility to be enrolled in the trial. Fourteen patients were excluded as they failed to fulfill the inclusion criteria (N = 6) or refused to participate in the trial (N = 8). Finally, 109 patients were randomly assigned to either Group 1, (N = 36), Group 2 (N = 37), or Control-group (N = 36) in a 1:1:1 ratio. Two doses of eltrombopag - 25 mg/day and 50 mg/day were administered to Group-1 and Group-2 patients, respectively whereas the control-group patients received standard dengue treatment without eltrombopag. The management of all enrolled patients was according to WHO guidelines. The randomization procedure was performed by using a computerized system (STATA Inc.). CBC and immature platelet fraction (IPF) were monitored from Day-0 to Day-7. Absolute immature platelet count (A-IPC) was calculated from PLT count and IPF for each patient. Aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels were measured on Day-0 and Day-4 and an Ultrasonogram (USG) of the abdomen was performed on Day-4 and Day-7 for each patient. The efficacy of eltrombopag as the primary outcome of the trial was investigated by the proportion of patients with recovered platelet count receiving eltrombopag with corrected platelet count (platelet count above the lower normal limit: 150 × 109/L) on Day-7 of the enrollment as compared to the Control-group. As the secondary outcomes, the reduction of bleeding tendency in response to eltrombopag as well as the safety of eltrombopag in dengue patients were assessed. The safety was evaluated in case of adverse events, liver function enzymes AST/ALT levels and USG. This trial is registered with the international clinical trial registry, number SLCTR/2019/037. RESULTS: A total of 101 patients including 77 DF and 24 DHF patients completed the trial as eight patients left the trial without completing the follow-up. Patients of the different groups were compared with respect to mean age (26±8, 30±10 and 30±9 years for, Group-1,-2 and Control-group, respectively) (p-value= 0.23) and basal PLT count (Group-1: 58±24 × 109; Group-2: 52±29 × 109 and control-group: 55±30 × 109) (p-value= 0.63). The mean PLT counts for Group-1 (332 × 109/L ± 92) and Group-2 (371 × 109/L ± 111) were significantly higher than control-group (194 × 109/L ± 96) on Day-7 (adjusted p-value= 1.15 × 10-06 for Group-1 vs. Control-group, and adjusted p-value= 1.82 × 10-08 for Group-2 vs. Control-group).). On Day-7, 91% of Group-1 (N = 30) and Group-2 (N = 32) patients who received eltrombopag achieved primary endpoint of PLT count above than lower normal limit (150 × 109/L) (Group-1: 91%, OR: 8.33, 95% CI: 2.11 to 32.80, p-value: 0.0024 and Group-2: 91%, OR: 8.89, 95% CI: 2.26 to 34.89, p-value: 0.0017) compared to 55% (N = 18) of control-group patients who did not receive eltrombopag. The bleeding manifestations for thirteen out of fourteen grade-II DHF patients were subsided within Day-7 who received eltrombopag, whereas four out of ten grade-II DHF patients with PLT counts lower than the lower normal limit in the control group showed intermittent bleeding symptoms throughout the trial period. Mean A-IPC but not IPF was significantly higher for eltrombopag-treated groups in comparison to the Control-group. The frequency of the most common adverse events (vomiting and diarrheal tendencies) was similar in the treated-and control-groups (N = 5, 15%, and N = 3, 9% for Group-1 and -2, respectively vs. N = 4, 12% in the Control-group). Ten (30%) patients of Group-1 and, fourteen (40%) patients of Group-2 showed increased AST (U/L) as opposed to nine patients (27%) in the Control-group. Increased ALT levels were observed for three (9%), nine (26%), and seven (21%) patients belonging to the Group-1, -2, and Control-group, respectively. PLT counts higher than the upper normal limit (450 × 109/L) on Day-7 were observed for seven patients who were administered the higher dose (50 mg/day) in contrast to the three patients receiving the lower dose (25 mg/day). USG reports did not show thrombosis events in any of the patients. INTERPRETATION: The trial revealed that the administration of eltrombopag in a short regimen for three days was efficacious to restore the PLT count in DF and DHF patients. The higher number of A-IPCs in eltrombopag treated patients underscored the possible mode of action of eltrombopag through stimulating megakaryopoiesis in dengue patients. The trial hints toward the positive effect of eltrombopag in the cessation of bleeding manifestation. Administration of the lower dose (25 mg/day) of eltrombopag was shown to be safer and equally efficacious to the higher dose (50 mg/day) in treating dengue-infected patients.

Predicting and Designing Epitope Ensemble Vaccines against HTLV-1.

The infection mechanism and pathogenicity of Human T-lymphotropic virus 1 (HTLV-1) are ambiguously known for hundreds of years. Our knowledge about this virus is recently emerging. The purpose of the study is to design a vaccine targeting the envelope glycoprotein, GP62, an outer membrane protein of HTLV-1 that has an increased number of epitope binding sites. Data collection, clustering and multiple sequence alignment of HTLV-1 glycoprotein B, variability analysis of envelope Glycoprotein GP62 of HTLV-1, population protection coverage, HLA-epitope binding prediction, and B-cell epitope prediction were performed to predict an effective vaccine. Among all the predicted peptides, ALQTGITLV and VPSSSTPL epitopes interact with three MHC alleles. The summative population protection coverage worldwide by these epitopes as vaccine candidates was found nearly 70%. The docking analysis revealed that ALQTGITLV and VPSSSTPL epitopes interact strongly with the epitope-binding groove of HLA-A*02:03, and HLA-B*35:01, respectively, as this HLA molecule was found common with which every predicted epitope interacts. Molecular dynamics simulations of the docked complexes show they form stable complexes. So, these potential epitopes might pave the way for vaccine development against HTLV-1.

A Dynamic Mathematical Modeling Revelation about the Impact of Vaccination on Hepatitis B Virus-induced Infection and Death Rate in Bangladesh.

AIM: Attainment of sustainable development goal (SDG) targets requires reducing the rate of new hepatitis B virus (HBV)-induced infection and mortality rate to 90% and 65%, respectively, by 2030. Therefore, it is important to investigate the feasibility of reducing the required rates of HBV-induced infection and death incidents at the current rate of vaccination coverage in Bangladesh. Moreover, factors influencing vaccination coverage like negative bias toward girls during immunization can affect the current vaccination program and ultimately hinder the efforts to reduce HBV-induced infection and death rates. To investigate the possibility of reducing HBV-induced infection and death rates with current vaccination coverage, we adopted mathematical molding-based approach. MATERIALS AND METHODS: We developed a mathematical model based on the susceptible-infectious-recovered model to simulate the HBV-induced infection in children under the age of five at three different vaccination rates: 80, 90, and 95%. Additionally the impact of current vaccination coverage was assessed on HBV-induced death rates in the future. Moreover, we took advantage of the mathematical model to investigate the impact of negative bias toward girls in vaccination program on HBV-induced infection and death rates. RESULTS: The model simulations revealed that 10% increase in the vaccination rate from 80 to 90% can potentially contribute to the significant lowering (around 40%) of HBV-induced infection rate among children. When increased by 5% of vaccination rate from 90 to 95%, the HBV-infection rate is likely to be decreased by another 22%. Likewise, 44% reduction in HBV-induced death rate in the future (2050 onward) can potentially be achieved by 10% increase in the current vaccination rate from 80 to 90%, whereas 5% increase in the current vaccination rate (90-95%) may lead to 24% further reduction of death rate. These results underscored the significant impact of vaccination in reducing HBV-induced infection among children and future death rates in adults. Moreover, at 90% vaccination coverage, the negative bias of vaccination toward girls contributes to an increase of 15 and 12% of HBV-induced infection and death rates, respectively, in female subjects compared to their male counterparts. CONCLUSION: The current vaccination coverage (80-90%) is further aggravated by untimely vaccination, dropouts from vaccination program, and negative bias toward girls in vaccination program. Therefore, if the current situation persists, it will not be possible to accomplish the required reduction in HBV-induced infection and death rates by 2030, according to the SDG guidelines. Moreover negative bias in the vaccination program may intensify the HBV-induced infection and death rates in the future. CLINICAL SIGNIFICANCE: In light of the mathematical model, we suggest that the vaccination coverage should be increased to 95% without any negative bias toward girls. To accomplish this, the concerning authorities must ensure timely and full completion of the HBV vaccine schedules, reducing dropouts from vaccination program, and lastly preventing negative bias toward girls to uplift vaccination coverage to more than 95% with gender equality. Without these strategies, the necessary reduction in the HBV-induced infection and death rates in Bangladesh may not be attained per SDG directives. HOW TO CITE THIS ARTICLE: Chakraborty S, Chakravorty R, Alam S, et al. A Dynamic Mathematical Modeling Revelation about the Impact of Vaccination on Hepatitis B Virus-induced Infection and Death Rate in Bangladesh. Euroasian J Hepato-Gastroenterol 2019;9(2):84-90.

Screening of Human Proteins for Fluoride and Aluminum Binding.

Previous studies showed that prolonged exposure to fluoride (F-) and aluminum (Al3+) ions is associated with numerous diseases including neurological disorders. They don't have any known biological function. But they can bind with proteins that interact with ions similar to them. Such unwanted interactions affect the normal biological function of the target proteins, as well as their downstream protein-protein interactions. Several studies show the detrimental effects posed by them including Alzheimer's disease. However, their target proteins have never been reported. Here, we have screened for the human protein targets subjected to F- and Al3+ interactions by using data-driven prediction tools. We have identified 20 different proteins that directly bind with them (10 interact with fluoride and 10 with aluminum). In addition, protein-protein interaction has been explored to find the proteins that indirectly interact with F- and Al3+. We have found 86 indirect targets for F- and 90 for Al3+. Furthermore, 19 common protein targets have been identified, including proteins (9 out of 19) associated with neurodegenerative disorders. However, wet lab experiments are beyond our scopes to validate the binding networks. Additional studies must be warranted.

Twin SVM-Based Classification of Alzheimer's Disease Using Complex Dual-Tree Wavelet Principal Coefficients and LDA.

Alzheimer's disease (AD) is a leading cause of dementia, which causes serious health and socioeconomic problems. A progressive neurodegenerative disorder, Alzheimer's causes the structural change in the brain, thereby affecting behavior, cognition, emotions, and memory. Numerous multivariate analysis algorithms have been used for classifying AD, distinguishing it from healthy controls (HC). Efficient early classification of AD and mild cognitive impairment (MCI) from HC is imperative as early preventive care could help to mitigate risk factors. Magnetic resonance imaging (MRI), a noninvasive biomarker, displays morphometric differences and cerebral structural changes. A novel approach for distinguishing AD from HC using dual-tree complex wavelet transforms (DTCWT), principal coefficients from the transaxial slices of MRI images, linear discriminant analysis, and twin support vector machine is proposed here. The prediction accuracy of the proposed method yielded up to 92.65 ± 1.18 over the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset, with a specificity of 92.19 ± 1.56 and sensitivity of 93.11 ± 1.29, and 96.68 ± 1.44 over the Open Access Series of Imaging Studies (OASIS) dataset, with a sensitivity of 97.72 ± 2.34 and specificity of 95.61 ± 1.67. The accuracy, sensitivity, and specificity achieved using the proposed method are comparable or superior to those obtained by various conventional AD prediction methods.