Post-hospitalization Short Versus Long Steroid Taper Strategies in Patients With Acute Severe Ulcerative Colitis: A Comparison of Clinical Outcomes.

Background: Ulcerative colitis (UC) is a chronic inflammatory colon disease characterized by relapsing flares and remission episodes. However, the optimal steroid tapering strategy in patients hospitalized for acute severe UC (ASUC) remains relatively unknown. We aim to examine the clinical outcomes in patients hospitalized for ASUC regarding variable prednisone taper regimens upon discharge. Methods: We retrospectively reviewed all adult patients admitted to our facility with ASUC between 2000 and 2022. Patients were divided into 2 groups based on the duration of steroid taper on discharge (< 6 and > 6 weeks). Patients who had colectomy at index admission were excluded from the analysis. The primary outcome was rehospitalization for ASUC within 6 months of index admission. Secondary outcomes included the need for colectomy, worsening endoscopic disease extent and/or severity during the follow-up period (6 months), and a composite outcome as a surrogate of worsening disease (defined as a combination of all products above). Two-sample t-tests and Pearson's chi-square tests were used to compare the means of continuous and categorical variables, respectively. Multivariate logistic regression analysis was performed to identify independent predictors for rehospitalization with ASUC. Results: A total of 215 patients (short steroid taper = 91 and long steroid taper = 124) were analyzed. A higher number of patients in the long steroid taper group had a longer disease duration since diagnosis and moderate-severe endoscopic disease activity (63.8 vs. 25.6 months, p < 0.0001, 46.8% vs. 23.1%, P = ≤ .05, respectively). Both groups had similar disease extent, prior biologic therapy, and the need for inpatient rescue therapy. At the 6-month follow-up, rates of rehospitalization with a flare of UC were comparable between the 2 groups (68.3% vs. 68.5%, P = .723). On univariate and multivariate logistic regression, escalation of steroid dose within four weeks of discharge (aOR 6.09, 95% CI: 1.82-20.3, P  = .003) was noted to be the only independent predictor for rehospitalization with ASUC. Conclusions: This is the first study comparing clinical outcomes between post-discharge steroid tapering regimens in hospitalized patients for ASUC. Both examined steroid taper regimens upon discharge showed comparable clinical results. Hence, we suggest a short steroid taper as a standard post-hospitalization strategy in patients following ASUC encounters. It is likely to enhance patient tolerability and reduce steroid-related adverse effects without adversely affecting outcomes.

Our Experience in the Surgical Management of Arterio-Venous Malformations of the head and neck.

Introduction: Vascular anomalies comprise a diverse group of abnormalities in blood vessel morphogenesis that usually occur prenatally. Arterio-venous malformations (AVMs) are rare congenital vascular lesions accounting for 1.5% of all vascular anomalies, with 50% of them occur in the oral and maxillo-facial regions. Treatment of large, complex vascular lesions is a serious challenge for patients and surgeons because it can cause disfigurement and massive haemorrhage, which may be spontaneous or the result of surgical intervention. Our study aimed to demonstrate surgical management of massive AVMs of the head and neck. Method: This retrospective study shows the treatment outcomes of 28 patients with massive maxillo-facial vascular malformations, who presented to our department for treatment from 1 January 2015 to 31 July 2022. Results: Twenty-eight patients with a mean age of 17.32 ± 12.21 years (women: 15, men: 13) were enrolled in the study. Diagnosis included extra cranial AVMs of the head and neck region. Treatment modalities, in isolation or combination, included angioembolisation procedure, sclerotherapy, and surgery. Conclusion: Management of AVMs is challenging owing to the replacement of normal tissue by the diseased ones and the high rate of recurrence. Hence, multi-modal approaches are needed for the effective restoration of tissues.

Optimizing multimodal feature selection using binary reinforced cuckoo search algorithm for improved classification performance.

Background: Feature selection is a vital process in data mining and machine learning approaches by determining which characteristics, out of the available features, are most appropriate for categorization or knowledge representation. However, the challenging task is finding a chosen subset of elements from a given set of features to represent or extract knowledge from raw data. The number of features selected should be appropriately limited and substantial to prevent results from deviating from accuracy. When it comes to the computational time cost, feature selection is crucial. A feature selection model is put out in this study to address the feature selection issue concerning multimodal. Methods: In this work, a novel optimization algorithm inspired by cuckoo birds' behavior is the Binary Reinforced Cuckoo Search Algorithm (BRCSA). In addition, we applied the proposed BRCSA-based classification approach for multimodal feature selection. The proposed method aims to select the most relevant features from multiple modalities to improve the model's classification performance. The BRCSA algorithm is used to optimize the feature selection process, and a binary encoding scheme is employed to represent the selected features. Results: The experiments are conducted on several benchmark datasets, and the results are compared with other state-of-the-art feature selection methods to evaluate the effectiveness of the proposed method. The experimental results demonstrate that the proposed BRCSA-based approach outperforms other methods in terms of classification accuracy, indicating its potential applicability in real-world applications. In specific on accuracy of classification (average), the proposed algorithm outperforms the existing methods such as DGUFS with 32%, MBOICO with 24%, MBOLF with 29%, WOASAT 22%, BGSA with 28%, HGSA 39%, FS-BGSK 37%, FS-pBGSK 42%, and BSSA 40%.

Gastric intestinal metaplasia and gastric cancer prevention: Watchful waiting.

Gastric intestinal metaplasia (GIM), a common histologic finding, is associated with increased risk of gastric cancer, and GIM associated with Helicobacter pylori infection is classified as an environmental metaplastic atrophic gastritis. Patients may be asymptomatic or present with various dyspeptic symptoms. Autoimmune metaplastic atrophic gastritis is a less common but important cause of chronic gastritis. The Correa cascade describes the evolution of precancerous mucosal changes that lead to development of GIM, with differentiation of 2 histologic types of GIM (complete and incomplete) and the consequences of each type. The risk of progression to malignancy is higher with incomplete GIM. It is also higher for those who immigrate from regions with a high incidence of H pylori infection to areas where the incidence is low. Guidelines regarding endoscopic management of GIM vary by geographic region.

Impact of Genetic Variations on Thromboembolic Risk in Saudis with Sickle Cell Disease.

BACKGROUND: Sickle cell disease (SCD) is a Mendelian disease characterized by multigenic phenotypes. Previous reports indicated a higher rate of thromboembolic events (TEEs) in SCD patients. A number of candidate polymorphisms in certain genes (e.g., FVL, PRT, and MTHFR) were previously reported as risk factors for TEEs in different clinical conditions. This study aimed to genotype these genes and other loci predicted to underlie TEEs in SCD patients. METHODOLOGY: A multi-center genome-wide association study (GWAS) involving Saudi SCD adult patients with a history of TEEs (n = 65) and control patients without TEE history (n = 285) was performed. Genotyping used the 10× Affymetrix Axiom array, which includes 683,030 markers. Fisher's exact test was used to generate p-values of TEE associations with each single-nucleotide polymorphism (SNP). The haplotype analysis software tool version 1.05, designed by the University of Göttingen, Germany, was used to identify the common inherited haplotypes. RESULTS: No association was identified between the targeted single-nucleotide polymorphism rs1801133 in MTHFR and TEEs in SCD (p = 0.79). The allele frequency of rs6025 in FVL and rs1799963 in PRT in our cohort was extremely low (<0.01); thus, both variants were excluded from the analysis as no meaningful comparison was possible. In contrast, the GWAS analysis showed novel genome-wide associations (p < 5 × 10-8) with seven signals; five of them were located on Chr 11 (rs35390334, rs331532, rs317777, rs147062602, and rs372091), one SNP on Chr 20 (rs139341092), and another on Chr 9 (rs76076035). The other 34 SNPs located on known genes were also detected at a signal threshold of p < 5 × 10-6. Seven of the identified variants are located in olfactory receptor family 51 genes (OR51B5, OR51V1, OR51A1P, and OR51E2), and five variants were related to family 52 genes (OR52A5, OR52K1, OR52K2, and OR52T1P). The previously reported association between rs5006884-A in OR51B5 and fetal hemoglobin (HbF) levels was confirmed in our study, which showed significantly lower levels of HbF (p = 0.002) and less allele frequency (p = 0.003) in the TEE cases than in the controls. The assessment of the haplotype inheritance pattern involved the top ten significant markers with no LD (rs353988334, rs317777, rs14788626882, rs49188823, rs139349992, rs76076035, rs73395847, rs1368823, rs8888834548, and rs1455957). A haplotype analysis revealed significant associations between two haplotypes (a risk, TT-AA-del-AA-ins-CT-TT-CC-CC-AA, and a reverse protective, CC-GG-ins-GG-del-TT-CC-TT-GG-GG) and TEEs in SCD (p = 0.024, OR = 6.16, CI = 1.34-28.24, and p = 0.019, OR = 0.33, CI = 0.13-0.85, respectively). CONCLUSIONS: Seven markers showed novel genome-wide associations; two of them were exonic variants (rs317777 in OLFM5P and rs147062602 in OR51B5), and less significant associations (p < 5 × 10-6) were identified for 34 other variants in known genes with TEEs in SCD. Moreover, two 10-SNP common haplotypes were determined with contradictory effects. Further replication of these findings is needed.

Deep residual-dense network based on bidirectional recurrent neural network for atrial fibrillation detection.

Atrial fibrillation easily leads to stroke, cerebral infarction and other complications, which will seriously harm the life and health of patients. Traditional deep learning methods have weak anti-interference and generalization ability. Therefore, we propose a new-fashioned deep residual-dense network via bidirectional recurrent neural network (RNN) model for atrial fibrillation detection. The combination of one-dimensional dense residual network and bidirectional RNN for atrial fibrillation detection simplifies the tedious feature extraction steps, and constructs the end-to-end neural network to achieve atrial fibrillation detection through data feature learning. Meanwhile, the attention mechanism is utilized to fuse the different features and extract the high-value information. The accuracy of the experimental results is 97.72%, the sensitivity and specificity are 93.09% and 98.71%, respectively compared with other methods.

Transcriptomic and Proteomic Analysis Reveals the Potential Role of RBMS1 in Adipogenesis and Adipocyte Metabolism.

Adipocytes play a critical role in maintaining a healthy systemic metabolism by storing and releasing energy in the form of fat and helping to regulate glucose and lipid levels in the body. Adipogenesis is the process through which pre-adipocytes are differentiated into mature adipocytes. It is a complex process involving various transcription factors and signaling pathways. The dysregulation of adipogenesis has been implicated in the development of obesity and metabolic disorders. Therefore, understanding the mechanisms that regulate adipogenesis and the factors that contribute to its dysregulation may provide insights into the prevention and treatment of these conditions. RNA-binding motif single-stranded interacting protein 1 (RBMS1) is a protein that binds to RNA and plays a critical role in various cellular processes such as alternative splicing, mRNA stability, and translation. RBMS1 polymorphism has been shown to be associated with obesity and type 2 diabetes, but the role of RBMS1 in adipose metabolism and adipogenesis is not known. We show that RBMS1 is highly expressed during the early phase of the differentiation of the murine adipocyte cell line 3T3-L1 and is significantly upregulated in the adipose tissue depots and adipocytes of high-fat-fed mice, implying a possible role in adipogenesis and adipose metabolism. Knockdown of RBMS1 in pre-adipocytes impacted the differentiation process and reduced the expression of some of the key adipogenic markers. Transcriptomic and proteomic analysis indicated that RBMS1 depletion affected the expression of several genes involved in major metabolic processes, including carbohydrate and lipid metabolism. Our findings imply that RBMS1 plays an important role in adipocyte metabolism and may offer novel therapeutic opportunity for metabolic disorders such as obesity and type 2 diabetes.

Osteo-NeT: An Automated System for Predicting Knee Osteoarthritis from X-ray Images Using Transfer-Learning-Based Neural Networks Approach.

Knee osteoarthritis is a challenging problem affecting many adults around the world. There are currently no medications that cure knee osteoarthritis. The only way to control the progression of knee osteoarthritis is early detection. Currently, X-ray imaging is a central technique used for the prediction of osteoarthritis. However, the manual X-ray technique is prone to errors due to the lack of expertise of radiologists. Recent studies have described the use of automated systems based on machine learning for the effective prediction of osteoarthritis from X-ray images. However, most of these techniques still need to achieve higher predictive accuracy to detect osteoarthritis at an early stage. This paper suggests a method with higher predictive accuracy that can be employed in the real world for the early detection of knee osteoarthritis. In this paper, we suggest the use of transfer learning models based on sequential convolutional neural networks (CNNs), Visual Geometry Group 16 (VGG-16), and Residual Neural Network 50 (ResNet-50) for the early detection of osteoarthritis from knee X-ray images. In our analysis, we found that all the suggested models achieved a higher level of predictive accuracy, greater than 90%, in detecting osteoarthritis. However, the best-performing model was the pretrained VGG-16 model, which achieved a training accuracy of 99% and a testing accuracy of 92%.

Identification of novel interacts partners of ADAR1 enzyme mediating the oncogenic process in aggressive breast cancer.

Triple-negative breast cancer (TNBC) subtype is characterized by aggressive clinical behavior and poor prognosis patient outcomes. Here, we show that ADAR1 is more abundantly expressed in infiltrating breast cancer (BC) tumors than in benign tumors. Further, ADAR1 protein expression is higher in aggressive BC cells (MDA-MB-231). Moreover, we identify a novel interacting partners proteins list with ADAR1 in MDA-MB-231, using immunoprecipitation assay and mass spectrometry. Using iLoop, a protein-protein interaction prediction server based on structural features, five proteins with high iloop scores were discovered: Histone H2A.V, Kynureninase (KYNU), 40S ribosomal protein SA, Complement C4-A, and Nebulin (ranged between 0.6 and 0.8). In silico analysis showed that invasive ductal carcinomas had the highest level of KYNU gene expression than the other classifications (p < 0.0001). Moreover, KYNU mRNA expression was shown to be considerably higher in TNBC patients (p < 0.0001) and associated with poor patient outcomes with a high-risk value. Importantly, we found an interaction between ADAR1 and KYNU in the more aggressive BC cells. Altogether, these results propose a new ADAR-KYNU interaction as potential therapeutic targeted therapy in aggressive BC.

Comparison between metabolic-associated fatty liver disease and nonalcoholic fatty liver disease: From nomenclature to clinical outcomes.

As a result of the obesity epidemic, Nonalcoholic fatty liver disease (NAFLD) and its complications have increased among millions of people. Consequently, a group of experts recommended changing the term NAFLD to an inclusive terminology more reflective of the underlying pathogenesis; metabolic-associated fatty liver disease (MAFLD). This new term of MAFLD has its own disease epidemiology and clinical outcomes prompting efforts in studying its differences from NAFLD. This article discusses the rationale behind the nomenclature change, the main differences, and its clinical implications.

Quantitative proteomics analysis of COVID-19 patients: Fetuin-A and tetranectin as potential modulators of innate immune responses.

Treatment of severe cases of coronavirus disease 2019 (COVID-19) is extremely important to minimize death and end-organ damage. Here we performed a proteomic analysis of plasma samples from mild, moderate and severe COVID-19 patients. Analysis revealed differentially expressed proteins and different therapeutic potential targets related to innate immune responses such as fetuin-A, tetranectin (TN) and paraoxonase-1 (PON1). Furthermore, protein changes in plasma showed dysregulation of complement and coagulation cascades in COVID-19 patients compared to healthy controls. In conclusion, our proteomics data suggested fetuin-A and TN as potential targets that might be used for diagnosis as well as signatures for a better understanding of the pathogenesis of COVID-19 disease.

Whole genome transcriptomic reveals heat stroke molecular signatures in humans.

An evolutionary heat shock response (HSR) protects most living species, including humans, from heat-induced macromolecular damage. However, its role in the pathogenesis of heat stroke is unknown. We examined the whole genome transcriptome in peripheral blood mononuclear cells of a cohort of subjects exposed to the same high environmental heat conditions, who developed heat stroke (n = 19) versus those who did not (n = 19). Patients with heat stroke had a mean rectal temperature at admission of 41.7 ± 0.8°C, and eight were in deep coma (Glasgow Coma Score = 3). The transcriptome showed that genes involved in more than half of the entire chaperome were differentially expressed relative to heat stress control. These include the heat shock protein, cochaperone, and chaperonin genes, indicating a robust HSR. Differentially expressed genes also encoded proteins related to unfolded protein response, DNA repair, energy metabolism, oxidative stress, and immunity. The analysis predicted perturbations of the proteome network and energy production. Cooling therapy attenuated these alterations without complete restoration of homeostasis. We validated the significantly expressed genes by a real-time polymerase chain reaction. The findings reveal the molecular signature of heat stroke. They also suggested that a powerful HSR may not be sufficient to protect against heat injury. The overwhelming proteotoxicity and energy failure could play a pathogenic role. KEY POINTS: Most living species, including humans, have inherent heat stress response (HSR) that shields them against heat-induced macromolecular damage. The role of the HSR in subjects exposed to environmental heat who progressed to heat stroke versus those that did not is unknown. Our findings suggest that heat stroke induces a broad and robust HSR of nearly half of the total heat shock proteins, cochaperones, and chaperonin genes. Heat stroke patients exhibited inhibition of genes involved in energy production, including oxidative phosphorylation and ATP production. Significant enrichment of neurodegenerative pathways, including amyloid processing signalling, the Huntington's and Parkinson's disease signalling suggestive of brain proteotoxicity was noted. The data suggests that more than a powerful HSR may be required to protect against heat stroke. Overwhelming proteotoxicity and energy failure might contribute to its pathogenesis.

Profiling of G-Protein Coupled Receptors in Adipose Tissue and Differentiating Adipocytes Offers a Translational Resource for Obesity/Metabolic Research.

G protein-coupled receptors (GPCRs) are expressed essentially on all cells, facilitating cellular responses to external stimuli, and are involved in nearly every biological process. Several members of this family play significant roles in the regulation of adipogenesis and adipose metabolism. However, the expression and functional significance of a vast number of GPCRs in adipose tissue are unknown. We used a high-throughput RT-PCR panel to determine the expression of the entire repertoire of non-sensory GPCRs in mouse white, and brown adipose tissue and assess changes in their expression during adipogenic differentiation of murine adipocyte cell line, 3T3-L1. In addition, the expression of GPCRs in subcutaneous adipose tissues from lean, obese, and diabetic human subjects and in adipocytes isolated from regular chow and high-fat fed mice were evaluated by re-analyzing RNA-sequencing data. We detected a total of 292 and 271 GPCRs in mouse white and brown adipose tissue, respectively. There is a significant overlap in the expression of GPCRs between the two adipose tissue depots, but several GPCRs are specifically expressed in one of the two tissue types. Adipogenic differentiation of 3T3-L1 cells had a profound impact on the expression of several GPCRs. RNA sequencing of subcutaneous adipose from healthy human subjects detected 255 GPCRs and obesity significantly changed the expression of several GPCRs in adipose tissue. High-fat diet had a significant impact on adipocyte GPCR expression that was similar to human obesity. Finally, we report several highly expressed GPCRs with no known role in adipose biology whose expression was significantly altered during adipogenic differentiation, and/or in the diseased human subjects. These GPCRs could play an important role in adipose metabolism and serve as a valuable translational resource for obesity and metabolic research.

Liver transplantation amidst the COVID-19 era: Our center's experience.

Coronavirus disease 2019 significantly impacted the liver transplant process worldwide. Consequently, it brought significant challenges and limitations to transplant policies and organ allocation forcing liver transplant centers to adjust their protocols to ensure maximum benefit and avoid harm to their patients. Our center, like many others, was obliged to adapt to the challenges. This paper provided an overview of the effects of coronavirus disease 2019 on liver transplantations and detailed our center's experience and efforts during this unprecedented pandemic to serve as a guide for future public health crises.

AI-enabled radiologist in the loop: novel AI-based framework to augment radiologist performance for COVID-19 chest CT medical image annotation and classification from pneumonia.

A SARS-CoV-2 virus-specific reverse transcriptase-polymerase chain reaction (RT-PCR) test is usually used to diagnose COVID-19. However, this test requires up to 2 days for completion. Moreover, to avoid false-negative outcomes, serial testing may be essential. The availability of RT-PCR test kits is currently limited, highlighting the need for alternative approaches for the precise and rapid diagnosis of COVID-19. Patients suspected to be infected with SARS-CoV-2 can be assessed using chest CT scan images. However, CT images alone cannot be used for ruling out SARS-CoV-2 infection because individual patients may exhibit normal radiological results in the primary phases of the disease. A machine learning (ML)-based recognition and segmentation system was developed to spontaneously discover and compute infection areas in CT scans of COVID-19 patients. The computable assessment exhibited suitable performance for automatic infection region allocation. The ML models developed were suitable for the direct detection of COVID-19 (+). ML was confirmed to be a complementary diagnostic technique for diagnosing COVID-19(+) by forefront medical specialists. The complete manual delineation of COVID-19 often requires up to 225.5 min; however, the proposed RILML method decreases the delineation time to 7 min after four iterations of model updating.

Transcriptome Changes and Metabolic Outcomes After Bariatric Surgery in Adults With Obesity and Type 2 Diabetes.

Context: Bariatric surgery has been shown to be effective in inducing complete remission of type 2 diabetes in adults with obesity. However, its efficacy in achieving complete diabetes remission remains variable and difficult to predict before surgery. Objective: We aimed to characterize bariatric surgery-induced transcriptome changes associated with diabetes remission and the predictive role of the baseline transcriptome. Methods: We performed a whole-genome microarray in peripheral mononuclear cells at baseline (before surgery) and 2 and 12 months after bariatric surgery in a prospective cohort of 26 adults with obesity and type 2 diabetes. We applied machine learning to the baseline transcriptome to identify genes that predict metabolic outcomes. We validated the microarray expression profile using a real-time polymerase chain reaction. Results: Sixteen patients entered diabetes remission at 12 months and 10 did not. The gene-expression analysis showed similarities and differences between responders and nonresponders. The difference included the expression of critical genes (SKT4, SIRT1, and TNF superfamily), metabolic and signaling pathways (Hippo, Sirtuin, ARE-mediated messenger RNA degradation, MSP-RON, and Huntington), and predicted biological functions (ß-cell growth and proliferation, insulin and glucose metabolism, energy balance, inflammation, and neurodegeneration). Modeling the baseline transcriptome identified 10 genes that could hypothetically predict the metabolic outcome before bariatric surgery. Conclusion: The changes in the transcriptome after bariatric surgery distinguish patients in whom diabetes enters complete remission from those who do not. The baseline transcriptome can contribute to the prediction of bariatric surgery-induced diabetes remission preoperatively.

Molecular Dynamics Simulations to Decipher the Role of Phosphorylation of SARS-CoV-2 Nonstructural Proteins (nsps) in Viral Replication.

Protein phosphorylation is a post-translational modification that enables various cellular activities and plays essential roles in protein interactions. Phosphorylation is an important process for the replication of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2). To shed more light on the effects of phosphorylation, we used an ensemble of neural networks to predict potential kinases that might phosphorylate SARS-CoV-2 nonstructural proteins (nsps) and molecular dynamics (MD) simulations to investigate the effects of phosphorylation on nsps structure, which could be a potential inhibitory target to attenuate viral replication. Eight target candidate sites were found as top-ranked phosphorylation sites of SARS-CoV-2. During the process of molecular dynamics (MD) simulation, the root-mean-square deviation (RMSD) analysis was used to measure conformational changes in each nsps. Root-mean-square fluctuation (RMSF) was employed to measure the fluctuation in each residue of 36 systems considered, allowing us to evaluate the most flexible regions. These analysis shows that there are significant structural deviations in the residues namely nsp1 THR 72, nsp2 THR 73, nsp3 SER 64, nsp4 SER 81, nsp4 SER 455, nsp5 SER284, nsp6 THR 238, and nsp16 SER 132. The identified list of residues suggests how phosphorylation affects SARS-CoV-2 nsps function and stability. This research also suggests that kinase inhibitors could be a possible component for evaluating drug binding studies, which are crucial in therapeutic discovery research.

Chest X ray and cough sample based deep learning framework for accurate diagnosis of COVID-19.

All witnessed the terrible effects of the COVID-19 pandemic on the health and work lives of the population across the world. It is hard to diagnose all infected people in real time since the conventional medical diagnosis of COVID-19 patients takes a couple of days for accurate diagnosis results. In this paper, a novel learning framework is proposed for the early diagnosis of COVID-19 patients using hybrid deep fusion learning models. The proposed framework performs early classification of patients based on collected samples of chest X-ray images and Coswara cough (sound) samples of possibly infected people. The captured cough samples are pre-processed using speech signal processing techniques and Mel frequency cepstral coefficient features are extracted using deep convolutional neural networks. Finally, the proposed system fuses extracted features to provide 98.70% and 82.7% based on Chest-X ray images and cough (audio) samples for early diagnosis using the weighted sum-rule fusion method.

ChromoEnhancer: An Artificial-Intelligence-Based Tool to Enhance Neoplastic Karyograms as an Aid for Effective Analysis.

Cytogenetics laboratory tests are among the most important procedures for the diagnosis of genetic diseases, especially in the area of hematological malignancies. Manual chromosomal karyotyping methods are time consuming and labor intensive and, hence, expensive. Therefore, to alleviate the process of analysis, several attempts have been made to enhance karyograms. The current chromosomal image enhancement is based on classical image processing. This approach has its limitations, one of which is that it has a mandatory application to all chromosomes, where customized application to each chromosome is ideal. Moreover, each chromosome needs a different level of enhancement, depending on whether a given area is from the chromosome itself or it is just an artifact from staining. The analysis of poor-quality karyograms, which is a difficulty faced often in preparations from cancer samples, is time consuming and might result in missing the abnormality or difficulty in reporting the exact breakpoint within the chromosome. We developed ChromoEnhancer, a novel artificial-intelligence-based method to enhance neoplastic karyogram images. The method is based on Generative Adversarial Networks (GANs) with a data-centric approach. GANs are known for the conversion of one image domain to another. We used GANs to convert poor-quality karyograms into good-quality images. Our method of karyogram enhancement led to robust routine cytogenetic analysis and, therefore, to accurate detection of cryptic chromosomal abnormalities. To evaluate ChromoEnahancer, we randomly assigned a subset of the enhanced images and their corresponding original (unenhanced) images to two independent cytogeneticists to measure the karyogram quality and the elapsed time to complete the analysis, using four rating criteria, each scaled from 1 to 5. Furthermore, we compared the enhanced images with our method to the original ones, using quantitative measures (PSNR and SSIM metrics).

An Efficient Machine Learning-Based Feature Optimization Model for the Detection of Dyslexia.

Dyslexia is among the most common neurological disorders in children. Detection of dyslexia therefore remains an important pursuit for the research works across various domains which is illustrated by the plethora of work presented in diverse scientific articles. The work presented herein attempted to utilize the potential of a unified gaming test of subjects (dyslexia/controls) in tandem with principal components derived from data to detect dyslexia. The work aims to build a machine learning model for dyslexia detection using comprehensive gaming test data. We have attempted to explore the potential of various kernel functions of the Support Vector Machine (SVM) on different number of principal components to reduce the computational complexity. A detection accuracy of 92% is obtained from the radial basis function with 5 components, and the highest detection accuracy obtained from the radial basis function with 3 components is 93%. On the contrary, the Artificial Neural Network(ANN) shows an added advantage with minimal number of hyperparameters with 3 components for obtaining an accuracy of 95%. The comparison of the proposed method with some of the existing works shows efficacy of this method for dyslexia detection.