Pub Date : 2023-02-01DOI: 10.1109/AICAPS57044.2023.10074336
Deepak S. Dharrao, A. Bongale, Vikrant Kadalaskar, Utkarsh Singh, Tathagata Singharoy
Text summarization is the process of extracting the meaning and important points from the text. It helps gain important information from the text while separating futile data. For generating a lot of textual data manually a person will be required to go through all the documents and then generate the summary which can be time taking and tiresome. Here Automatic text summarization (ATS) comes into the picture which takes text as input and generates the summary of that text with the help of machine learning algorithms and natural language processing techniques or NLP techniques. The use of ATS in the medical field can help doctors go through a patient’s medical history in a shorter period of time and take better decisions about the diagnosis of the patient.
摘要是从文本中提取意义和要点的过程。它有助于从文本中获得重要信息，同时分离无用的数据。为了手动生成大量文本数据，需要一个人浏览所有文档，然后生成摘要，这既耗时又令人厌烦。自动文本摘要(Automatic text summarization, ATS)是一种将文本作为输入，并在机器学习算法和自然语言处理技术或NLP技术的帮助下生成文本摘要的方法。ATS在医疗领域的应用可以帮助医生在更短的时间内了解患者的病史，并对患者的诊断做出更好的决定。
{"title":"Patients’ Medical History Summarizer using NLP","authors":"Deepak S. Dharrao, A. Bongale, Vikrant Kadalaskar, Utkarsh Singh, Tathagata Singharoy","doi":"10.1109/AICAPS57044.2023.10074336","DOIUrl":"https://doi.org/10.1109/AICAPS57044.2023.10074336","url":null,"abstract":"Text summarization is the process of extracting the meaning and important points from the text. It helps gain important information from the text while separating futile data. For generating a lot of textual data manually a person will be required to go through all the documents and then generate the summary which can be time taking and tiresome. Here Automatic text summarization (ATS) comes into the picture which takes text as input and generates the summary of that text with the help of machine learning algorithms and natural language processing techniques or NLP techniques. The use of ATS in the medical field can help doctors go through a patient’s medical history in a shorter period of time and take better decisions about the diagnosis of the patient.","PeriodicalId":146698,"journal":{"name":"2023 International Conference on Advances in Intelligent Computing and Applications (AICAPS)","volume":"172 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120872115","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-01DOI: 10.1109/AICAPS57044.2023.10074518
G. Sreeraag, P. Shynu
Social networks have had a significant impact on people's personal and professional life all around the world. Since the COVID-19 pandemic has boosted the use of digital media among people, fake news and reviews have had a stronger impact on society in recent years. This study demonstrates how the stiffness index may be used to model the spread of fake news in Indian states. We demonstrate that the speed at which fake news circulates through online social networks increases with a stiffness index. We conducted a stiffness analysis for all Indian states to assess the spread of fake information in each Indian state. The stiffness analysis of the conventional SIR model, one of the widely used approaches to describe the propagation of rumors in social networks, serves as an explanation and illustration of our proposition. The rise in fake news in our society is also justified by a comparison of the stiffness index for India before and after the COVID-19 outbreak. The study provides governments and policymakers with a more comprehensive understanding of the value of early intervention to combat the spread of false information via digital media.
{"title":"Stiffness Analysis for the Prediction of Fake News through Online Digital Networks in India","authors":"G. Sreeraag, P. Shynu","doi":"10.1109/AICAPS57044.2023.10074518","DOIUrl":"https://doi.org/10.1109/AICAPS57044.2023.10074518","url":null,"abstract":"Social networks have had a significant impact on people's personal and professional life all around the world. Since the COVID-19 pandemic has boosted the use of digital media among people, fake news and reviews have had a stronger impact on society in recent years. This study demonstrates how the stiffness index may be used to model the spread of fake news in Indian states. We demonstrate that the speed at which fake news circulates through online social networks increases with a stiffness index. We conducted a stiffness analysis for all Indian states to assess the spread of fake information in each Indian state. The stiffness analysis of the conventional SIR model, one of the widely used approaches to describe the propagation of rumors in social networks, serves as an explanation and illustration of our proposition. The rise in fake news in our society is also justified by a comparison of the stiffness index for India before and after the COVID-19 outbreak. The study provides governments and policymakers with a more comprehensive understanding of the value of early intervention to combat the spread of false information via digital media.","PeriodicalId":146698,"journal":{"name":"2023 International Conference on Advances in Intelligent Computing and Applications (AICAPS)","volume":"101 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121212121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-01DOI: 10.1109/AICAPS57044.2023.10074283
Deepak S. Dharrao, S. Gite, Rahee Walambe
Inverse Reinforcement Learning is a subset of Imitation learning, where the goal is to generate a reward function that captures an expert’s behavior using a set of demonstrations by the expert. Guided Cost Learning (GCL) is a recent approach to finding a neural network reward function. In this paper the GCL algorithm is explored and applied to the Lunar Lander environment of the OpenAI gym. We generated our own set of expert demonstrations and implemented the GCL algorithm. We successfully demonstrate that Guided Cost Learning can generate a reward that completely encapsulates desired behavior depicted in the expert demonstrations, even for high dimensional state space environments such as the lunar lander environment. Reward and policy evaluations between the actual reward function and the GCL generated rewards function are compared and the results are presented.
{"title":"Guided Cost Learning for Lunar Lander Environment Using Human Demonstrated Expert Trajectories","authors":"Deepak S. Dharrao, S. Gite, Rahee Walambe","doi":"10.1109/AICAPS57044.2023.10074283","DOIUrl":"https://doi.org/10.1109/AICAPS57044.2023.10074283","url":null,"abstract":"Inverse Reinforcement Learning is a subset of Imitation learning, where the goal is to generate a reward function that captures an expert’s behavior using a set of demonstrations by the expert. Guided Cost Learning (GCL) is a recent approach to finding a neural network reward function. In this paper the GCL algorithm is explored and applied to the Lunar Lander environment of the OpenAI gym. We generated our own set of expert demonstrations and implemented the GCL algorithm. We successfully demonstrate that Guided Cost Learning can generate a reward that completely encapsulates desired behavior depicted in the expert demonstrations, even for high dimensional state space environments such as the lunar lander environment. Reward and policy evaluations between the actual reward function and the GCL generated rewards function are compared and the results are presented.","PeriodicalId":146698,"journal":{"name":"2023 International Conference on Advances in Intelligent Computing and Applications (AICAPS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122555829","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-01DOI: 10.1109/AICAPS57044.2023.10074121
Ananya, Rishabh Kaushal
News bulletins play an important role in people’s daily lives. As humans evolved, so did our ability to form opinions. In the domain of journalism and news reporting, it is desirable that those reporting news do not add their personal opinions. However, we often observe biases in news reporting, and therefore the task of assessing the opinion of news reporters has become a significant issue. In this work, we study the performance of classical machine learning and vectorization techniques on opinion detection in Hinglish code-mixed news debates related to political and religious issues aired on Indian news channels. We were able to achieve the best accuracy of 87% using Logistic Regression algorithm with Bag of Words (BoW) vectorization technique.
{"title":"Opinion Detection in Hinglish News Reporting","authors":"Ananya, Rishabh Kaushal","doi":"10.1109/AICAPS57044.2023.10074121","DOIUrl":"https://doi.org/10.1109/AICAPS57044.2023.10074121","url":null,"abstract":"News bulletins play an important role in people’s daily lives. As humans evolved, so did our ability to form opinions. In the domain of journalism and news reporting, it is desirable that those reporting news do not add their personal opinions. However, we often observe biases in news reporting, and therefore the task of assessing the opinion of news reporters has become a significant issue. In this work, we study the performance of classical machine learning and vectorization techniques on opinion detection in Hinglish code-mixed news debates related to political and religious issues aired on Indian news channels. We were able to achieve the best accuracy of 87% using Logistic Regression algorithm with Bag of Words (BoW) vectorization technique.","PeriodicalId":146698,"journal":{"name":"2023 International Conference on Advances in Intelligent Computing and Applications (AICAPS)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121644418","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-01DOI: 10.1109/AICAPS57044.2023.10074488
Meet Kumari
A hybrid visible light communication (VLC)-fiber link is an favorable selection in various types of geographical restriction areas from urban to rural areas for advanced internet of things (IoT) applications. It helps in reduce the cost of overall system at high data rate as well as long reach communication. In this work, a red light emitting diode (LED) based fiber-VLC system has been designed. The comparative analysis of using four LEDs in the proposed work reveals that red LED based VLC-fiber transmission system offer high data rate of 30Gbps. It also provides faithful fiber range of 40km and 80m VLC range under the presence of noise. Besides this, the proposed system is a superior system as compared to other related work.
{"title":"Design of IoT based hybrid Red LED VLC-fiber communication system","authors":"Meet Kumari","doi":"10.1109/AICAPS57044.2023.10074488","DOIUrl":"https://doi.org/10.1109/AICAPS57044.2023.10074488","url":null,"abstract":"A hybrid visible light communication (VLC)-fiber link is an favorable selection in various types of geographical restriction areas from urban to rural areas for advanced internet of things (IoT) applications. It helps in reduce the cost of overall system at high data rate as well as long reach communication. In this work, a red light emitting diode (LED) based fiber-VLC system has been designed. The comparative analysis of using four LEDs in the proposed work reveals that red LED based VLC-fiber transmission system offer high data rate of 30Gbps. It also provides faithful fiber range of 40km and 80m VLC range under the presence of noise. Besides this, the proposed system is a superior system as compared to other related work.","PeriodicalId":146698,"journal":{"name":"2023 International Conference on Advances in Intelligent Computing and Applications (AICAPS)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123636873","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-01DOI: 10.1109/AICAPS57044.2023.10074198
G. K, V. S.
Tuberculosis (TB) is an infectious illness that may be severe and primarily impacts the lungs. Examining sputum smears under bright field microscopes is one of the simplest and most successful ways to detect TB infection in impoverished nations like India. A method for detecting tuberculosis bacteria from bright-field microscopic sputum smear images is proposed in this work. U-shaped encoder-decoder network architecture (U-Net) is used to first segment the bright field microscopic sputum smear images, and then Random Forest Classification Algorithm is used for final prediction. The detection of bacilli produced results that are comparable to other methods.
{"title":"Identification of Tuberculosis Bacilli from Bright Field Microscopic Sputum Smear Images using U-Net and Random Forest Classification Algorithm","authors":"G. K, V. S.","doi":"10.1109/AICAPS57044.2023.10074198","DOIUrl":"https://doi.org/10.1109/AICAPS57044.2023.10074198","url":null,"abstract":"Tuberculosis (TB) is an infectious illness that may be severe and primarily impacts the lungs. Examining sputum smears under bright field microscopes is one of the simplest and most successful ways to detect TB infection in impoverished nations like India. A method for detecting tuberculosis bacteria from bright-field microscopic sputum smear images is proposed in this work. U-shaped encoder-decoder network architecture (U-Net) is used to first segment the bright field microscopic sputum smear images, and then Random Forest Classification Algorithm is used for final prediction. The detection of bacilli produced results that are comparable to other methods.","PeriodicalId":146698,"journal":{"name":"2023 International Conference on Advances in Intelligent Computing and Applications (AICAPS)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133057061","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-01DOI: 10.1109/AICAPS57044.2023.10074030
D. L, C. R
In the world we live in today, massive amounts of data are transferred in a matter of seconds. Internet of Medical Things (IoMT) is a technology that enables the health parameters of patients to be collected by medical devices and transmitted through Internet to a remote server for analysis by medical experts. This highly sensitive data can be affected by malware which causes threats to human lives. In this scenario, the application of Artificial Intelligent techniques have high impact on the analysis of malignancy in the health parameters. Boosting algorithms are very efficient in the classification of data. This paper proposes an EXtreme Gradient Boosting algorithm (XGBoost) for the detection of malware present in the data. The hyperparameters of the XGB algorithm are optimised using an intelligent evolutionary technique named as Differential Evolution (DE) . The experiment is conducted on a WUSTL EHMS 2020 Dataset for Internet of Medical Things (IoMT) CyberSecurity dataset and produced an accuracy of 97.39% after hyperparameter optimisation. The DE optimised XGB Classifier performed well in the detection of malware with regard to accuracy and speed.
在我们今天生活的世界里，大量的数据在几秒钟内被传输。医疗物联网(Internet of Medical Things, IoMT)是指通过医疗设备收集患者的健康参数，并通过互联网传输到远程服务器，供医疗专家进行分析的技术。这些高度敏感的数据可能会受到恶意软件的影响，从而对人类生命造成威胁。在这种情况下，人工智能技术的应用对健康参数的恶性分析有很大的影响。增强算法在数据分类方面非常有效。本文提出了一种用于检测数据中存在的恶意软件的极限梯度增强算法(XGBoost)。XGB算法的超参数使用一种称为差分进化(DE)的智能进化技术进行优化。实验在WUSTL EHMS 2020医疗物联网(IoMT)网络安全数据集上进行，经过超参数优化，准确率达到97.39%。DE优化的XGB分类器在检测恶意软件的准确性和速度方面表现良好。
{"title":"An Optimal Differential Evolution Based XGB Classifier for IoMT malware classification","authors":"D. L, C. R","doi":"10.1109/AICAPS57044.2023.10074030","DOIUrl":"https://doi.org/10.1109/AICAPS57044.2023.10074030","url":null,"abstract":"In the world we live in today, massive amounts of data are transferred in a matter of seconds. Internet of Medical Things (IoMT) is a technology that enables the health parameters of patients to be collected by medical devices and transmitted through Internet to a remote server for analysis by medical experts. This highly sensitive data can be affected by malware which causes threats to human lives. In this scenario, the application of Artificial Intelligent techniques have high impact on the analysis of malignancy in the health parameters. Boosting algorithms are very efficient in the classification of data. This paper proposes an EXtreme Gradient Boosting algorithm (XGBoost) for the detection of malware present in the data. The hyperparameters of the XGB algorithm are optimised using an intelligent evolutionary technique named as Differential Evolution (DE) . The experiment is conducted on a WUSTL EHMS 2020 Dataset for Internet of Medical Things (IoMT) CyberSecurity dataset and produced an accuracy of 97.39% after hyperparameter optimisation. The DE optimised XGB Classifier performed well in the detection of malware with regard to accuracy and speed.","PeriodicalId":146698,"journal":{"name":"2023 International Conference on Advances in Intelligent Computing and Applications (AICAPS)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115566218","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-01DOI: 10.1109/AICAPS57044.2023.10074504
Blessy Paul P, Cini Kurian
In recent years, the number of online courses in India has skyrocketed especially due to the Covid pandemic. The most significant increments have happened in degree colleges, where 85% concur that internet based courses are important for their drawn-out procedure when contrasted with 60% in 2015. The distribution of online courses has evolved dramatically as technology has advanced. Web-based platform provides new challenges for both teachers and students. Teachers should be clear about the effectiveness of online learning in teaching students. For that, the possibilities of online learning should be compared with traditional learning. Students are evaluated based on their focus on online learning. This study aims to determine the efficacy of online courses by predicting student performance in an e-learning system. These research findings evaluate modern learning methods, highlight students’ potential and help teachers understand how to assess and lead students on online platforms.
{"title":"Student performance prediction in e-learning system and evaluating effectiveness of online courses","authors":"Blessy Paul P, Cini Kurian","doi":"10.1109/AICAPS57044.2023.10074504","DOIUrl":"https://doi.org/10.1109/AICAPS57044.2023.10074504","url":null,"abstract":"In recent years, the number of online courses in India has skyrocketed especially due to the Covid pandemic. The most significant increments have happened in degree colleges, where 85% concur that internet based courses are important for their drawn-out procedure when contrasted with 60% in 2015. The distribution of online courses has evolved dramatically as technology has advanced. Web-based platform provides new challenges for both teachers and students. Teachers should be clear about the effectiveness of online learning in teaching students. For that, the possibilities of online learning should be compared with traditional learning. Students are evaluated based on their focus on online learning. This study aims to determine the efficacy of online courses by predicting student performance in an e-learning system. These research findings evaluate modern learning methods, highlight students’ potential and help teachers understand how to assess and lead students on online platforms.","PeriodicalId":146698,"journal":{"name":"2023 International Conference on Advances in Intelligent Computing and Applications (AICAPS)","volume":"98 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120906475","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-01DOI: 10.1109/AICAPS57044.2023.10074389
Yasir Altaf, Abdul Wahid
Convolutional neural networks (CNNs) have been widely used in hand gesture classification problems, and have made a major contribution to this area by overcoming the limitations of hard-code feature extraction techniques. CNN in hand gesture classification aims to improve performance through automatic feature engineering. Several researchers have used various CNN architectures to accurately classify hand gestures.In this paper, we investigate the performance of a popular CNN variant called dilated CNN to classify hand gestures into their corresponding classes. We compared the performance of the dilated CNN with that of the standard CNN on two benchmark ISL and ASL datasets. The experimental results demonstrate that the dilated CNN significantly enhances performance compared to the standard CNN. We obtained a significant increase in accuracy for both datasets using the dilated-CNN compared to the standard CNN.
{"title":"Evaluation of Dilated CNN for Hand Gesture Classification","authors":"Yasir Altaf, Abdul Wahid","doi":"10.1109/AICAPS57044.2023.10074389","DOIUrl":"https://doi.org/10.1109/AICAPS57044.2023.10074389","url":null,"abstract":"Convolutional neural networks (CNNs) have been widely used in hand gesture classification problems, and have made a major contribution to this area by overcoming the limitations of hard-code feature extraction techniques. CNN in hand gesture classification aims to improve performance through automatic feature engineering. Several researchers have used various CNN architectures to accurately classify hand gestures.In this paper, we investigate the performance of a popular CNN variant called dilated CNN to classify hand gestures into their corresponding classes. We compared the performance of the dilated CNN with that of the standard CNN on two benchmark ISL and ASL datasets. The experimental results demonstrate that the dilated CNN significantly enhances performance compared to the standard CNN. We obtained a significant increase in accuracy for both datasets using the dilated-CNN compared to the standard CNN.","PeriodicalId":146698,"journal":{"name":"2023 International Conference on Advances in Intelligent Computing and Applications (AICAPS)","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125981068","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-01DOI: 10.1109/AICAPS57044.2023.10074363
Kishankumar Vaishnani, Bakul Gohel, Avik Hati
Cell nuclei count and morphology are the key parameters in the histopathological image for evaluating various pathological conditions. However, the manual extraction of these parameters is a tedious and time-consuming task. Automated nuclei segmentation is the practical solution. Deep learning-based approaches have recently become popular for automated nuclei segmentation tasks in histopathological images. Stain colour variability frequently occurs in Hematoxylin and Eosin (H&E)-stained histopathological images because of differences in the staining process and digitisation medium. A deep learning-based approach is susceptible to data variability; therefore, data augmentation and normalisation are crucial pre-processing steps to improve the model's generalisation. In the present work, we performed the comparative analysis of the colour augmentation and stain normalisation techniques, namely Reinhard, Macenko and Vahadane, for deep learning-based nuclei segmentation tasks in H&E stained histopathological images. We have used three different datasets and performed within-dataset and cross- dataset analysis to evaluate the trained model's generalisation capabilities. Stain normalisation methods, e.g. Reinhard and Vahadane, showed better performance in various datasets than colour augmentation techniques.
{"title":"Impact of Stain Normalisation Technique on Deep Learning based Nuclei Segmentation in Histopathological Image","authors":"Kishankumar Vaishnani, Bakul Gohel, Avik Hati","doi":"10.1109/AICAPS57044.2023.10074363","DOIUrl":"https://doi.org/10.1109/AICAPS57044.2023.10074363","url":null,"abstract":"Cell nuclei count and morphology are the key parameters in the histopathological image for evaluating various pathological conditions. However, the manual extraction of these parameters is a tedious and time-consuming task. Automated nuclei segmentation is the practical solution. Deep learning-based approaches have recently become popular for automated nuclei segmentation tasks in histopathological images. Stain colour variability frequently occurs in Hematoxylin and Eosin (H&E)-stained histopathological images because of differences in the staining process and digitisation medium. A deep learning-based approach is susceptible to data variability; therefore, data augmentation and normalisation are crucial pre-processing steps to improve the model's generalisation. In the present work, we performed the comparative analysis of the colour augmentation and stain normalisation techniques, namely Reinhard, Macenko and Vahadane, for deep learning-based nuclei segmentation tasks in H&E stained histopathological images. We have used three different datasets and performed within-dataset and cross- dataset analysis to evaluate the trained model's generalisation capabilities. Stain normalisation methods, e.g. Reinhard and Vahadane, showed better performance in various datasets than colour augmentation techniques.","PeriodicalId":146698,"journal":{"name":"2023 International Conference on Advances in Intelligent Computing and Applications (AICAPS)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114793899","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：