ENTAIL: yEt aNoTher amyloid fIbrils cLassifier

Citarella, Alessia Auriemma; Biasi, Luigi Di; Marco, Fabiola De; Tortora, Genoveffa

doi:10.1186/s12859-022-05070-6

Cited by 12 publications

(5 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With an accuracy on the test set of 81.80%, a sensibility of 100%, and a specificity of 63.63% on a balanced dataset, ENTAIL was based on the Naive Bayes Classifier with Unbounded Support and Gaussian Kernel Type. The investigation that was carried out showed that performance is superior in terms of performance on a balanced dataset [ 139 ] despite the various configurations of the tests.…”

Section: Computational Methods For Amyloid Fibril Identificationmentioning

confidence: 99%

Machine Learning Approaches in Diagnosis, Prognosis and Treatment Selection of Cardiac Amyloidosis

Allegra

Mirabile

Tonacci

et al. 2023

IJMS

View full text Add to dashboard Cite

Cardiac amyloidosis is an uncommon restrictive cardiomyopathy featuring an unregulated amyloid protein deposition that impairs organic function. Early cardiac amyloidosis diagnosis is generally delayed by indistinguishable clinical findings of more frequent hypertrophic diseases. Furthermore, amyloidosis is divided into various groups, according to a generally accepted taxonomy, based on the proteins that make up the amyloid deposits; a careful differentiation between the various forms of amyloidosis is necessary to undertake an adequate therapeutic treatment. Thus, cardiac amyloidosis is thought to be underdiagnosed, which delays necessary therapeutic procedures, diminishing quality of life and impairing clinical prognosis. The diagnostic work-up for cardiac amyloidosis begins with the identification of clinical features, electrocardiographic and imaging findings suggestive or compatible with cardiac amyloidosis, and often requires the histological demonstration of amyloid deposition. One approach to overcome the difficulty of an early diagnosis is the use of automated diagnostic algorithms. Machine learning enables the automatic extraction of salient information from “raw data” without the need for pre-processing methods based on the a priori knowledge of the human operator. This review attempts to assess the various diagnostic approaches and artificial intelligence computational techniques in the detection of cardiac amyloidosis.

show abstract

Section: Computational Methods For Amyloid Fibril Identificationmentioning

confidence: 99%

Machine Learning Approaches in Diagnosis, Prognosis and Treatment Selection of Cardiac Amyloidosis

Allegra

Mirabile

Tonacci

et al. 2023

IJMS

View full text Add to dashboard Cite

show abstract

“…Notably, the chosen datasets underscore the issue of class imbalance, where one dataset (CHSLB) exhibits balanced classes (51% and 49%), while six out of eight datasets have one class representing over 60% of the samples. In the Kidney-Abu Dhabi dataset, one class comprises a striking 89% of the data, highlighting the significance of addressing this common challenge for accurate predictions in medical contexts [83], [84]. In addressing the inherent imbalance within the datasets, the proposed methodology incorporated a combination of strategic adjustments within the XGBoost framework.…”

Section: Resultsmentioning

confidence: 99%

Synergistic Feature Engineering and Ensemble Learning for Early Chronic Disease Prediction

Al-Jamimi

2024

IEEE Access

View full text Add to dashboard Cite

Chronic diseases, a global public health challenge, necessitate the deployment of cutting-edge predictive models for early diagnosis and personalized interventions. This study presents an advanced methodology for early prediction of chronic diseases, including heart attack, diabetes, breast cancer, and kidney disease, leveraging a synergistic combination of cutting-edge techniques. Recognizing the challenge posed by extensive medical datasets with numerous features, we introduce a novel approach that begins with Feature Engineering using Recursive Feature Elimination (RFE) in conjunction with a Support Vector Machine (SVM). The presented methodology identifies and removes irrelevant features to simplify data complexity. The refined dataset is then input into the robust eXtreme Gradient Boosting (XGBoost) classifier, known for its efficiency and adeptness in predicting complex relationships within the data. The chosen ensemble learning algorithm demonstrates significant prowess in inducing intricate patterns crucial for chronic disease prediction. To enhance model performance, an essential phase of optimization is introduced through hyperparameter tuning using Bayesian optimization. This strategically navigates the hyperparameter space, maximizing the efficiency of the search process and fine-tuning the model for optimal predictive accuracy. The proposed approach showcases a substantial improvement in the early prediction of chronic diseases, demonstrating the effectiveness of the proposed approach.

show abstract

“…[9]. When it comes to the performance of these algorithms, the use of a balanced medical dataset results in good performance [10], [11].…”

Section: Related Workmentioning

confidence: 99%

Knee Osteoarthritis Analysis Using Deep Learning and XAI on X-Rays

Ahmed,

Imran

2024

IEEE Access

View full text Add to dashboard Cite

Knee osteoarthritis (OA) is a chronic disorder mainly arising from age-related factors affecting the knee joints. Its diagnosis is critically important which is usually done by doctors using X-ray images. Although this process is accurate, it is time-consuming. X-ray images have facilitated the use of deep learning (DL) models for the automation of the diagnosis of knee OA, commonly employing convolutional neural network (CNN) based architectures. However, the lack of models' interpretability makes the results less trustworthy. This work builds on the existing state-of-the-art (SOTA) pre-trained DL models to understand the model's behavior in classifying highly complex knee OA cases utilizing a divide-and-conquer approach -from multi-class to a binary class for better results interpretability and explainability using explainable artificial intelligence (XAI). Five SOTA fine-tuned DL models are tested on Kellgren-Lawrence (KL) graded X-ray images. Both multi-class and binary-class (using the multiple subsets derived from the original dataset to examine how the models perform with different data combinations) classification approaches and their interpretability of findings using Gradient-weighted Class Activation Mapping (GradCAM) are undertaken in this study. The GradCAM visualization of EfficientNetb7 demonstrates that when the degree of variance between different classes increases, the model's efficiency in classifying knee OA also increases. Specifically, it becomes more effective at distinguishing normal and severe cases with 99.13% classification accuracy. However, the model's efficacy drops to 67% for other cases, indicating that it cannot classify knee OA as effectively as doctors.INDEX TERMS Explainable Artificial Intelligence (XAI), deep learning, knee osteoarthritis, healthcare, diagnosis, and classification.

show abstract

ENTAIL: yEt aNoTher amyloid fIbrils cLassifier

Cited by 12 publications

References 33 publications

Machine Learning Approaches in Diagnosis, Prognosis and Treatment Selection of Cardiac Amyloidosis

Machine Learning Approaches in Diagnosis, Prognosis and Treatment Selection of Cardiac Amyloidosis

Synergistic Feature Engineering and Ensemble Learning for Early Chronic Disease Prediction

Knee Osteoarthritis Analysis Using Deep Learning and XAI on X-Rays

Contact Info

Product

Resources

About