Related references
Note: Only part of the references are listed.
Article
Medicine, General & Internal
Yasmin Mohd Yacob et al.
Summary: Atrophic gastritis (AG) is a chronic condition caused by H. pylori infection and can lead to gastric cancer if left untreated. Early detection of AG is crucial to prevent such cases.
Article
Medicine, General & Internal
Ala'a Zyout et al.
Summary: This study used two different spectrum representations, iris-spectrogram and scalogram, to extract deep features and classify different ECG beat waves using two deep convolutional neural networks (CNN), ResNet101 and ShuffleNet. The results showed that using ResNet101 and scalogram of T-wave achieved the highest accuracy of 98.3% for beat rhythm detection, while using iris-spectrogram and ResNet101 for QRS-wave achieved an accuracy of 94.4%. In conclusion, deep features from time-frequency representation of ECG beat waves can accurately detect basic rhythms such as normal, tachycardia, and bradycardia.
Article
Medicine, General & Internal
Riaz Ahmad et al.
Summary: This paper presents an improved hybrid approach for efficient white blood cell (WBC) subtype classification. It uses transfer learning on pre-trained deep neural networks to extract optimum deep features from enhanced and segmented WBC images, and then filters the feature vector using an entropy-controlled marine predator algorithm. The proposed method achieves an overall average accuracy of 99.9% with more than 95% reduction in the size of the feature vector.
Article
Medicine, General & Internal
Hatim Butt et al.
Summary: Diabetes patients can access personalized glycemic profiles for efficient prediction of future blood glucose levels. This research proposes a transformation method for event-based data and develops a multi-layered LSTM-based recurrent neural network to predict blood glucose levels in type 1 diabetes patients. The proposed method achieves the lowest RMSE scores of 14.76 mg/dL and 25.48 mg/dL for 30 min and 60 min prediction horizons, respectively. The results can be utilized in closed-loop systems for precise insulin delivery to improve glycemic control in type 1 patients.
Article
Medicine, General & Internal
Abdul Aziz K. Abdul Hamid et al.
Summary: Improving the accuracy and efficiency of time-series forecasts is crucial for authorities to predict and prevent the spread of the Coronavirus disease. The dataset contains both linear and non-linear patterns, which makes it inefficient to use linear models for prediction. A hybrid approach is proposed to achieve a more accurate and efficient predictive value of COVID-19.
Review
Medicine, General & Internal
Marina Yusoff et al.
Summary: Breast cancer diagnosis relies on histopathological imaging, which is time-consuming due to image complexity and volume. Deep learning has become popular for diagnosing cancerous images but achieving high precision and minimizing overfitting remains challenging for classification solutions.
Article
Medicine, General & Internal
Mehmet Erten et al.
Summary: A novel text-based classification model has been developed to discriminate between SARS-CoV-2 and Influenza-A infections, showing high accuracy in screening protein sequences.
Article
Medicine, General & Internal
Keijiro Nakamura et al.
Summary: This study developed and validated a deep learning-based prognostic model to predict the risk of all-cause mortality for patients with HF. The proposed model showed better prediction performance in terms of discrimination, calibration, and risk stratification compared to other deep learning and traditional statistical models, especially in identifying high-risk patients.
Article
Medicine, General & Internal
Naseha Wafa Qammar et al.
Summary: This study employs the notion of perfect matrices of Lagrange differences to detect atrial fibrillation episodes based on three ECG parameters. The results show that the sensitivity of algebraic relationships between cardiac intervals increases when the dimension of the perfect matrices of Lagrange differences is extended. By establishing a baseline dataset and utilizing a decision support system for classification, it can help determine whether new candidates have indications of atrial fibrillation. Probability distribution graphs and semi-gauge indicator techniques aid in visualizing the categorization of the new candidates.
Article
Medicine, General & Internal
Abdelkader Dairi et al.
Summary: This paper introduces an unsupervised deep learning-driven scheme for recognizing mental tasks using EEG signals. The proposed scheme applies the Multichannel Wiener filter for robust recognition and a quadratic time-frequency distribution (QTFD) for effective signal representation. The features extracted by QTFD are then classified using a deep belief network (DBN)-driven Isolation Forest (iF) detector. Experimental results show that the proposed DBN-based iF detector achieves superior discrimination performance compared to other EEG-based classification methods.
Article
Medicine, General & Internal
Ahmed Almulihi et al.
Summary: This paper proposes a deep stacking ensemble model to enhance the performance of heart disease prediction. The proposed model integrates two optimized and pre-trained hybrid deep learning models with Support Vector Machine as the meta-learner model, and uses Recursive Feature Elimination for feature selection optimization. The proposed model has been tested on two different heart disease datasets and achieved the highest performance.
Article
Medicine, General & Internal
Parvathaneni Naga Srinivasu et al.
Summary: The development of genomic technology has become the most demanding area for computer-aided diagnostic and treatment research. Advances in machine learning and artificial intelligence have allowed for the prediction of future illnesses and the development of accurate models. This research focuses on predicting type 2 diabetes using gene sequences and testing the performance of neural network models.