Multi-Agent System Based on Machine Learning for Early Diagnosis of Diabetes

Chakour, Imane; Mourabit, Yousef El; Daoui, Cherki; Baslam, Mohamed

doi:10.1109/icoa49421.2020.9094511

Cited by 8 publications

(4 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Data quality, privacy concerns, and algorithmic biases demand careful consideration. Responsible AI deployment in healthcare is pivotal, and ethical guidelines are emerging to ensure fairness, bias mitigation, and data privacy [25]. Future research directions encompass the refinement of algorithms, multi-modal data integration, and the development of explainable AI for healthcare, signifying the dynamic nature of this field.…”

Section: Discussionmentioning

confidence: 99%

Health Prediction Web Application Using Flask Framework

2024

IRJMETS

View full text Add to dashboard Cite

In the realm of proactive healthcare, disease prediction through machine learning has emerged as a transformative force. The survey commences by tracing the evolution of machine learning in healthcare, exploring the historical context, and tracking the exponential growth of this field. Subsequently, a diverse range of machine learning algorithms are employed for disease prediction, categorizing them into supervised learning, unsupervised learning, and other emerging techniques. The review comprehensively evaluates the strengths and limitations of each algorithm, shedding light on their applications across various medical domains. There is also underscore on the importance of privacy-preserving techniques in health data analysis, emphasising the necessity of data security and user confidentiality. Ethical considerations, fairness, bias mitigation, and responsible deployment of machine learning in healthcare feature prominently in this discourse. However, as with any frontier, there are challenges. Furthermore, this survey is forward-looking, identifying unexplored research avenues and promising future directions in disease prediction through machine learning. In this synthesis of knowledge, there is not only distilled the essence of the field but also bring to the fore exemplary case studies, showcasing notable projects in disease prediction. These real-world case studies offer practical insights into methodologies, results, and contributions, enriching our understanding of the subject.

show abstract

Section: Discussionmentioning

confidence: 99%

Health Prediction Web Application Using Flask Framework

2024

IRJMETS

View full text Add to dashboard Cite

show abstract

“…In Reference (8), they created a new multi-agent system for diabetes diagnosis at an early stage. Three classifiers included in this model as ANN, SVM, and LR.…”

Section: Related Workmentioning

confidence: 99%

Analyzing Three Predictive Algorithms for Diabetes Mellitus Against the Pima Indians Dataset

Darolia¹,

Chhillar²

2022

ECS Trans.

View full text Add to dashboard Cite

Diabetes mellitus is a metabolic disease in which the pancreas fails to produce enough insulin required for the processing of blood glucose. Most medical institutions analyze Electronic Health Records (EHRs) manually and then predict whether the patient is diabetic or not. The objective of this work is to classify diabetes and non-diabetes patients using predictive algorithms/techniques. These algorithms provide cost, time, and effort-effective solutions for the prognosis and diagnosis of diabetes mellitus. In this work, popular algorithms like Artificial Neural Network (ANN), Random Forest (RF), and Logistic Regression (LR) have been used against the PIMA Indians. Dataset and analysis has been carried out on open-source software WEKA. In addition, this paper provided state-of-the-art by various researchers related to the said topic. This work concluded that LR outperforms other algorithms, with the accuracy of 77.10%, but in the case of area under the curve (0.83), both LR and RF perform equally well.

show abstract

“…However, other approaches focus on distributed data resource, which is important and needed in many applications due to the cost of data transformation, limited network bandwidth, data restriction issues, and privacy concerns. In distributed data environments, distributed learning success lies in minimal information transferring, low communication and computation costs, and models combining [5]. Various distributed machine learning methods were proposed to build a general model for distributed sites and attempted to solve several distributed data resources issues.…”

Section: Introductionmentioning

confidence: 99%

Decentralised and Privacy Preserving Machine Learning for Multiple Distributed Data Resources

Alkhozae

Zeng

2021

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

There are many issues related to distributed data resources, such as data transfer restrictions, privacy, and computation and communication costs. It brings challenges for developing a distributed machine learning approach that can overcome these issues without using complicated machine learning methods, sharing, or centralising the distributed data resources. To address these issues, we proposed a decentralised machine learning approach that allows distributed sites to develop classification models at global and local levels without sharing or moving distributed data to a centralised database or using a central server for iterative computation or communication. Such a decentralised approach restricts the exchanged information between distributed sites with only the trained local models and minimal information. The experimental evaluation conducted on different datasets reported that our approach performed better or close to the centralised or other existing distributed machine learning approaches. Also, we applied a model selection strategy based on F-measure and different combination methods. That is, the proposed method can be used to distributed data resources efficiently and preserve data privacy without sacrificing the predictive performance.

show abstract

Multi-Agent System Based on Machine Learning for Early Diagnosis of Diabetes

Cited by 8 publications

References 2 publications

Health Prediction Web Application Using Flask Framework

Health Prediction Web Application Using Flask Framework

Analyzing Three Predictive Algorithms for Diabetes Mellitus Against the Pima Indians Dataset

Decentralised and Privacy Preserving Machine Learning for Multiple Distributed Data Resources

Contact Info

Product

Resources

About