Shobhit Agarwal
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Shobhit Agarwal has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
- Girish Kumar Pati
Professor, Department of Gastroenterology, Institute of Medical Sciences and Sum Hospital, Bhubaneswar, India
2 shared publications - Jimmy Narayan
Department of Gastroenterology, Institute of Medical Sciences and SUM Hospital, Bhubaneswar, India
2 shared publications Ayaskant Singh
Department of Gastroenterology
1 shared publication- Ayaskanta Singh
SUM Ultimate Medicare, Bhubaneswar, India
1 shared publication Chinmayee Kar
IMS and SUM Hospital; Bhubaneswar, Odisha, India
1 shared publicationDebakanta Mishra
Department of Gastroenterology
1 shared publication
Latest Publications
- Open Access
- 31 March 2023
- Published by Springer Nature
- p. 73-94
A Study on Data Augmentation Techniques for Visual Defect Detection in Manufacturing
Deep learning-based defect detection is rapidly gaining importance for automating visual quality control tasks in industrial applications. However, due to usually low rejection rates in manufacturing processes, industrial defect detection datasets are inherent to three severe data challenges: data sparsity, data imbalance, and data shift. Because the acquisition of defect data is highly cost″​=intensive, and Deep Learning (DL) algorithms require a sufficiently large amount of data, we are investigating how to solve these challenges using data oversampling and data augmentation (DA) techniques. Given the problem of binary defect detection, we present a novel experimental procedure for analyzing the impact of different DA-techniques. Accordingly, pre-selected DA-techniques are used to generate experiments across multiple datasets and DL models. For each defect detection use-case, we configure a set of random DA-pipelines to generate datasets of different characteristics. To investigate the impact of DA-techniques on defect detection performance, we then train convolutional neural networks with two different but fixed architectures and hyperparameter sets. To quantify and evaluate the generalizability, we compute the distances between dataset derivatives to determine the degree of domain shift. The results show that we can precisely analyze the influences of individual DA-methods, thus laying the foundation for establishing a mapping between dataset properties and DA-induced performance enhancement aiming for enhancing DL development. We show that there is no one-fits all solution, but that within the categories of geometrical and color augmentations, certain DA-methods outperform others.
- 31 March 2023
- Published by Springer Nature
- p. 73-94
Open AccessA Study on Data Augmentation Techniques for Visual Defect Detection in Manufacturing
Deep learning-based defect detection is rapidly gaining importance for automating visual quality control tasks in industrial applications. However, due to usually low rejection rates in manufacturing processes, industrial defect detection datasets are inherent to three severe data challenges: data sparsity, data imbalance, and data shift. Because the acquisition of defect data is highly cost″​=intensive, and Deep Learning (DL) algorithms require a sufficiently large amount of data, we are investigating how to solve these challenges using data oversampling and data augmentation (DA) techniques. Given the problem of binary defect detection, we present a novel experimental procedure for analyzing the impact of different DA-techniques. Accordingly, pre-selected DA-techniques are used to generate experiments across multiple datasets and DL models. For each defect detection use-case, we configure a set of random DA-pipelines to generate datasets of different characteristics. To investigate the impact of DA-techniques on defect detection performance, we then train convolutional neural networks with two different but fixed architectures and hyperparameter sets. To quantify and evaluate the generalizability, we compute the distances between dataset derivatives to determine the degree of domain shift. The results show that we can precisely analyze the influences of individual DA-methods, thus laying the foundation for establishing a mapping between dataset properties and DA-induced performance enhancement aiming for enhancing DL development. We show that there is no one-fits all solution, but that within the categories of geometrical and color augmentations, certain DA-methods outperform others.
- Open Access
- 2 December 2021
- Published by Wiley in International Journal of Hepatology
- Vol. 2021, 1-8
Diastolic Dysfunction Is a Predictor of Poor Survival in Patients with Decompensated Cirrhosis
Background. Left ventricular diastolic dysfunction (LVDD) appears to be the earliest cardiac disturbance in cirrhosis patients. There are many previous reports reporting the significance of severity of LVDD on the outcome of liver transplantation or TIPS insertion, a few Indian studies have addressed the role of LVDD on survival in decompensated cirrhosis. The objective of this study is to assess the effect of LVDD on the survival of decompensated cirrhotic patients. Methods. We prospectively evaluated 92 decompensated cirrhotic patients from April 2015 to March 2017 at IMS and SUM Hospital, Bhubaneswar, India. 2D echocardiography with tissue Doppler imaging was used to evaluate cardiac function, as per the American society of echocardiography guidelines. The primary endpoint was to evaluate the effect of LVDD on overall mortality. Results. Ninety-two decompensated cirrhotic patients were evaluated in this prospective cohort study. Twenty-eight out of 92 patients (30%) died due to liver-related complications after a follow-up of 24 months. The decompensated cirrhotic patients with had a significantly higher ratio ( vs. , ) suggesting severe LV dysfunction in advanced cirrhosis. Patients with had significantly higher MELD score and Child-Pugh score ( vs. ; vs.
- 2 December 2021
- Published by Wiley in International Journal of Hepatology
- Vol. 2021, 1-8
Open AccessDiastolic Dysfunction Is a Predictor of Poor Survival in Patients with Decompensated Cirrhosis
Background. Left ventricular diastolic dysfunction (LVDD) appears to be the earliest cardiac disturbance in cirrhosis patients. There are many previous reports reporting the significance of severity of LVDD on the outcome of liver transplantation or TIPS insertion, a few Indian studies have addressed the role of LVDD on survival in decompensated cirrhosis. The objective of this study is to assess the effect of LVDD on the survival of decompensated cirrhotic patients. Methods. We prospectively evaluated 92 decompensated cirrhotic patients from April 2015 to March 2017 at IMS and SUM Hospital, Bhubaneswar, India. 2D echocardiography with tissue Doppler imaging was used to evaluate cardiac function, as per the American society of echocardiography guidelines. The primary endpoint was to evaluate the effect of LVDD on overall mortality. Results. Ninety-two decompensated cirrhotic patients were evaluated in this prospective cohort study. Twenty-eight out of 92 patients (30%) died due to liver-related complications after a follow-up of 24 months. The decompensated cirrhotic patients with had a significantly higher ratio ( vs. , ) suggesting severe LV dysfunction in advanced cirrhosis. Patients with had significantly higher MELD score and Child-Pugh score ( vs. ; vs.
- Open Access
- 12 April 2021
- Published by Springer Nature in Cureus
- Vol. 13  (4)
SPINK1 Mutation in Idiopathic Chronic Pancreatitis: How Pertinent Is It in Coastal Eastern India?
Background and aim Idiopathic chronic pancreatitis (ICP) is said to be present when no identifiable etiology can be identified. Robust evidence suggested that the serine protease inhibitor nucleus Kazol type 1 (SPINK1) N34S mutation was frequently associated with ICP. As there is a paucity of data on genetic studies in ICP cases from the coastal eastern region of India, we performed this study with an aim to evaluate the SPINK1 genetic mutations and other associated clinical correlates in ICP cases. Material and methods Consecutive ICP cases attending the department of gastroenterology, Institute of Medical Sciences (IMS) and SUM Hospital, were enrolled and evaluated for the pertinent clinical history and undergone detailed biochemical and radiological evaluations. Two ml of venous blood in ethylenediaminelelraacelic acid (EDTA) vials were collected from each case and subjected to a polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) test for genetic analysis. Result In this study, the mean age of the cases at the time of the first consultation with us and the age of the first clinical presentation were 34.5246.44 and 28.7345.52 years, respectively. Males outnumbered females (Male:Female - 2.12:1). Out of the total of 200 cases, 50% had no SPINK1 mutation, whereas 40% and 10% cases had SPINK I N34S heterozygous and homozygous mutations, respectively. The mean age of clinical presentation, severe abdominal pain, exocrine and endocrine insufficiency, and parenchymal atrophy were significantly more common in mutants as compared to non-mutants (p-value <0.05). Conclusion In our region, 50% of ICP cases had the SPINK1 N34S mutation. The SPINK1 mutants had a relatively more severe variety of pancreatitis as compared to non-mutants.
- 12 April 2021
- Published by Springer Nature in Cureus
- Vol. 13  (4)
Open AccessSPINK1 Mutation in Idiopathic Chronic Pancreatitis: How Pertinent Is It in Coastal Eastern India?
Background and aim Idiopathic chronic pancreatitis (ICP) is said to be present when no identifiable etiology can be identified. Robust evidence suggested that the serine protease inhibitor nucleus Kazol type 1 (SPINK1) N34S mutation was frequently associated with ICP. As there is a paucity of data on genetic studies in ICP cases from the coastal eastern region of India, we performed this study with an aim to evaluate the SPINK1 genetic mutations and other associated clinical correlates in ICP cases. Material and methods Consecutive ICP cases attending the department of gastroenterology, Institute of Medical Sciences (IMS) and SUM Hospital, were enrolled and evaluated for the pertinent clinical history and undergone detailed biochemical and radiological evaluations. Two ml of venous blood in ethylenediaminelelraacelic acid (EDTA) vials were collected from each case and subjected to a polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) test for genetic analysis. Result In this study, the mean age of the cases at the time of the first consultation with us and the age of the first clinical presentation were 34.5246.44 and 28.7345.52 years, respectively. Males outnumbered females (Male:Female - 2.12:1). Out of the total of 200 cases, 50% had no SPINK1 mutation, whereas 40% and 10% cases had SPINK I N34S heterozygous and homozygous mutations, respectively. The mean age of clinical presentation, severe abdominal pain, exocrine and endocrine insufficiency, and parenchymal atrophy were significantly more common in mutants as compared to non-mutants (p-value <0.05). Conclusion In our region, 50% of ICP cases had the SPINK1 N34S mutation. The SPINK1 mutants had a relatively more severe variety of pancreatitis as compared to non-mutants.
- 15 September 2020
- Published by Egyptian Knowledge Bank in Microbes and Infectious Diseases
Liver Cirrhosis and Concomitant Gastric Helicobacter Pylori Infection
Background: Cirrhosis of liver causes a lot of morbidities and mortalities. Around one third Indians had Helicobacter Pylori (H.Pylori) infection, whose effect in disease course of cirrhosis is poorly defined. Therefore this study was undertaken with an aim to find out variable effects of Helicobacter Pylori infection in cirrhotic patients. Methods: This cross-sectional, observational, open labeled, non interventional, single centered study was carried out in the Department of Gastroenterology of Institute of Medical Sciences (I.M.S.) & SUM Hospital, Bhubaneswar in between January 2018 and December 2019. Consecutive cirrhotic cases attending the hospital were enrolled in the study, after undergoing biochemical, radiological and endoscopic evaluation. All of them also subjected to rapid urease test (RUT) to rule out Helicobacter Pylori infection. Results: Total 864 cirrhotic cases were included in this study, amongst which male outnumbered female. Mean age of presentation was 48.24±10.77 years. Alcohol was the most common etiology. Around 57.4% of cirrhosis cases had Helicobacter Pylori infection. 70.96% of alcoholic liver disease (ALD) and 50% of cryptogenic cirrhosis cases had Helicobacter Pylori infection, whereas none of chronic Hepatitis B virus (HBV) related cirrhotic had rapid urease test (RUT) positivity. Cases with Helicobacter Pylori infection presented early and had relatively higher prevalence of higher grade of Esophageal Varix, Portal hypertensive Gastropathy, Duodenal ulcer, Gastric Antral Vascular Ectasia, Gastroesophageal Varix II compared to cases without Helicobacter Pylori infection. The prognostic score such as model for end stage liver disease (MELD) was relatively higher in cases without Helicobacter Pylori infection compared to cases with Helicobacter Pylori infection. Conclusion: In our study, we found most of cirrhotic cases had alcoholic liver disease and were male. Around half of cases had Helicobacter Pylori infection and earlier presentation. Although the cases with Helicobacter Pylori infection had relatively higher endoscopic severity but had lower prognostic score compared to cases without Helicobacter Pylori infection, which should be validated in future by further studies.
- 15 September 2020
- Published by Egyptian Knowledge Bank in Microbes and Infectious Diseases
Liver Cirrhosis and Concomitant Gastric Helicobacter Pylori Infection
Background: Cirrhosis of liver causes a lot of morbidities and mortalities. Around one third Indians had Helicobacter Pylori (H.Pylori) infection, whose effect in disease course of cirrhosis is poorly defined. Therefore this study was undertaken with an aim to find out variable effects of Helicobacter Pylori infection in cirrhotic patients. Methods: This cross-sectional, observational, open labeled, non interventional, single centered study was carried out in the Department of Gastroenterology of Institute of Medical Sciences (I.M.S.) & SUM Hospital, Bhubaneswar in between January 2018 and December 2019. Consecutive cirrhotic cases attending the hospital were enrolled in the study, after undergoing biochemical, radiological and endoscopic evaluation. All of them also subjected to rapid urease test (RUT) to rule out Helicobacter Pylori infection. Results: Total 864 cirrhotic cases were included in this study, amongst which male outnumbered female. Mean age of presentation was 48.24±10.77 years. Alcohol was the most common etiology. Around 57.4% of cirrhosis cases had Helicobacter Pylori infection. 70.96% of alcoholic liver disease (ALD) and 50% of cryptogenic cirrhosis cases had Helicobacter Pylori infection, whereas none of chronic Hepatitis B virus (HBV) related cirrhotic had rapid urease test (RUT) positivity. Cases with Helicobacter Pylori infection presented early and had relatively higher prevalence of higher grade of Esophageal Varix, Portal hypertensive Gastropathy, Duodenal ulcer, Gastric Antral Vascular Ectasia, Gastroesophageal Varix II compared to cases without Helicobacter Pylori infection. The prognostic score such as model for end stage liver disease (MELD) was relatively higher in cases without Helicobacter Pylori infection compared to cases with Helicobacter Pylori infection. Conclusion: In our study, we found most of cirrhotic cases had alcoholic liver disease and were male. Around half of cases had Helicobacter Pylori infection and earlier presentation. Although the cases with Helicobacter Pylori infection had relatively higher endoscopic severity but had lower prognostic score compared to cases without Helicobacter Pylori infection, which should be validated in future by further studies.