The slicing based spreading analysis for melanoma prediction using reinforcement learning model DOI

Venkata Ashok K Gorantla,

Shiva Kumar Sriramulugari,

Amit Hasmukhbhai Mewada

et al.

Published: Dec. 14, 2023

the present study proposes a novel approach to skin lesion prediction, namely, slicing-based spreading analysis (SBSA) with reinforcement learning (RL) model. The aim of SBSA approach, as implemented in this study, is mine and capture key aspects data from different perspectives for more accurate classification. We additionally introduce RL models enhanced performance classification tasks. Specifically, our based on five phases: obtaining complete data, slicing collected repeating promotional process, training slices RL, finally, combining trained predicting type. A benchmark dataset 400 dermoscopic pictures was used test suggested melanoma identification. accuracy attained compared traditional like support vector machines (SVM), random forests (RF), multilayer perceptions (MLP) utilizing methodology. Results indicated that achieved better metrics than classic machine approaches. Furthermore, proposed models, an overall 94.56%, significantly outperforming other models. In conclusion, provides promising type prediction.

Language: Английский

Improved Skin Cancer Detection Using Efficient NetB7 Machine Learning DOI
Muskan Sharma, Yash Mahajan, Priya Batta

et al.

Published: Nov. 23, 2023

Early identification may help prevent or cure skin cancer, a serious worldwide health problem. To increase the accuracy and effectiveness of cancer prediction, this research employs machine learning approaches. The HAM10000 dataset was used to train test our brand-new prediction model, which is based on state-of-the-art EfficientNetB7 architecture. resolve class imbalance concerns that are prevalent in dermatological datasets, data augmentation procedures employed give equitable representation training data. RGB attributes were taken from photographs incorporated into model. Our approach outperforms conventional models with an 89%, promising. In addition improving algorithms, provides fast affordable option aid early detection. demonstrates value accurate detection patient outcomes reducing healthcare expenditures, as well potential diagnostics. This not only paves way for future developments area automated diagnosis but also offers hope wider applications medical image analysis.

Language: Английский

Citations

1
Privacy in the Machine Learning: A Study on User Profiling and Targeted Advertising on LinkedIn DOI Creative Commons

Eric UWAYEZU

Published: May 14, 2024

Abstract As the digital landscape changes, privacy concerns in machine learning applications need to be focused on. This research will investigate implications of LinkedIn platform related targeted advertising and user profiling. The main purpose this is understand algorithm used by generate profiles way they provide relevant users. use different methods, like interviews, surveys, data analysis. first step look at algorithms processes for collecting To what kind collected how create profiles, evaluate level control users have over their data. In process gathering information, surveys done on concern awareness platform's policies. A sample given interviews get more qualitative feedback users' experiences. check types are that keep them engaged with platform. study give a great picture taken advantage platform, from perspective there trade-off between content end, another catalyst huge conversation happening now giving new suggestions industry best practices improve findings open discussion ways, itself legislators

Language: Английский

Citations

0
Classification of Cipher Text by Clustering of S-Topological Rough Group DOI Creative Commons

D. Keerthana,

V. Visalakshi,

Prasanalakshmi Balaji

et al.

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 119302 - 119313

Published: Jan. 1, 2024

Language: Английский

Citations

0
Skin Cancer Classification with Channel Attention and SMOTE Sampling: A Deep Learning Approach DOI

Mohammad Imtiaj Mahbub,

Sk. Md. Masudul Ahsan

2019 4th International Conference on Electrical Information and Communication Technology (EICT), Journal Year: 2023, Volume and Issue: unknown, P. 1 - 6

Published: Dec. 7, 2023

Skin cancer is a prevalent and sometimes lifethreatening disease, with early detection being crucial for successful treatment. Dermatoscopic images have become valuable tool diagnosing skin lesions. In this study, machine learning techniques, especially deep learning, shown promise in automating the diagnosis process. This paper presents learning-based approach lesion classification aimed at enhancing accuracy reducing burden on healthcare professionals. The model architecture integrates conventional CNN layers novel Channel Attention Layer, which adaptively weights features extracted from different channels. enhancement allows to concentrate most informative elements of images, potentially leading better performance. To address challenge class imbalance dermatoscopic datasets, Synthetic Minority Over-sampling Technique (SMOTE) applied balance dataset while avoiding information loss. technique useful when dealing medical where certain types are less common. results study show an outstanding overall 92.47%. indicates efficacy proposed assist dermatologists precise timely diagnosis, consequently improving patient outcomes.

Language: Английский

Citations

0
Federated Learning for Secure Healthcare Image Analysis in the Cloud DOI
Neeraj Varshney, Parul Madan, Anurag Shrivastava

et al.

Published: Dec. 29, 2023

This study investigates the use of federated learning in healthcare picture analysis with goal improving diagnostic precision while safeguarding patient data privacy. A specialized framework was created, showing considerable gains precision, privacy protection, as well computational effectiveness. Sophisticated security measures, such access limits and encryption, successfully protected private medical data. Blockchain technology addition to suggested hybrid cloud architecture offered scalable secure alternatives for organizations. Decision-makers can take action based on practical ramifications. Future research ought concentrate customizing particular imaging modalities, investigating edge computing applications, evaluating long-term advantages difficulties field healthcare.

Language: Английский

Citations

0
The slicing based spreading analysis for melanoma prediction using reinforcement learning model DOI

Venkata Ashok K Gorantla,

Shiva Kumar Sriramulugari,

Amit Hasmukhbhai Mewada

et al.

Published: Dec. 14, 2023

the present study proposes a novel approach to skin lesion prediction, namely, slicing-based spreading analysis (SBSA) with reinforcement learning (RL) model. The aim of SBSA approach, as implemented in this study, is mine and capture key aspects data from different perspectives for more accurate classification. We additionally introduce RL models enhanced performance classification tasks. Specifically, our based on five phases: obtaining complete data, slicing collected repeating promotional process, training slices RL, finally, combining trained predicting type. A benchmark dataset 400 dermoscopic pictures was used test suggested melanoma identification. accuracy attained compared traditional like support vector machines (SVM), random forests (RF), multilayer perceptions (MLP) utilizing methodology. Results indicated that achieved better metrics than classic machine approaches. Furthermore, proposed models, an overall 94.56%, significantly outperforming other models. In conclusion, provides promising type prediction.

Language: Английский

Citations

0