Deep integrated fusion of local and global features for cervical cell classification

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 171, P. 108153 - 108153

Published: Feb. 14, 2024

Cervical cytology image classification is of great significance to the cervical cancer diagnosis and prognosis. Recently, convolutional neural network (CNN) visual transformer have been adopted as two branches learn features for by simply adding local global features. However, such simple addition may not be effective integrate these In this study, we explore synergy images tasks. Specifically, design a Deep Integrated Feature Fusion (DIFF) block synergize from CNN branch branch. Our proposed method evaluated on three cell datasets (SIPaKMeD, CRIC, Herlev) another large blood dataset BCCD several multi-class binary Experimental results demonstrate effectiveness in classification, which could assist medical specialists better diagnose cancer.

Language: Английский

---

RMS-UNet: Residual multi-scale UNet for liver and lesion segmentation

Artificial Intelligence in Medicine, Journal Year: 2022, Volume and Issue: 124, P. 102231 - 102231

Published: Jan. 12, 2022

Language: Английский

---

Deep learning algorithm performance evaluation in detection and classification of liver disease using CT images

Multimedia Tools and Applications, Journal Year: 2023, Volume and Issue: 83(1), P. 2773 - 2790

Published: May 15, 2023

Language: Английский

---

A multi-modal deep neural network for multi-class liver cancer diagnosis

Neural Networks, Journal Year: 2023, Volume and Issue: 165, P. 553 - 561

Published: June 13, 2023

Language: Английский

---

Wavelet radiomics features from multiphase CT images for screening hepatocellular carcinoma: analysis and comparison

Scientific Reports, Journal Year: 2023, Volume and Issue: 13(1)

Published: Nov. 10, 2023

Early detection of liver malignancy based on medical image analysis plays a crucial role in patient prognosis and personalized treatment. This task, however, is challenging due to several factors, including data scarcity limited training samples. paper presents study three important aspects radiomics feature from multiphase computed tomography (CT) for classifying hepatocellular carcinoma (HCC) other focal lesions: wavelet-transformed extraction, relevant selection, features-based classification under the inadequate Our shows that combining features extracted wavelet original CT domains enhance performance significantly, compared with using those or domain only. To facilitate multi-domain combination, we introduce logistic sparsity-based model selection Bayesian optimization find proposed yields more discriminative than existing methods, filter-based, wrapper-based, model-based techniques. In addition, present comparison recent deep convolutional neural network (CNN)-based models hepatic lesion diagnosis. The results show scenario, produces comparable, if not higher, metrics CNN-based terms area curve.

Language: Английский

---

Machine learning-enabled healthcare information systems in view of Industrial Information Integration Engineering

Journal of Industrial Information Integration, Journal Year: 2022, Volume and Issue: 30, P. 100382 - 100382

Published: July 30, 2022

Language: Английский

---

An efficient classification of cirrhosis liver disease using hybrid convolutional neural network-capsule network

Biomedical Signal Processing and Control, Journal Year: 2022, Volume and Issue: 80, P. 104152 - 104152

Published: Oct. 14, 2022

Language: Английский

---

Adaptive Method for Exploring Deep Learning Techniques for Subtyping and Prediction of Liver Disease

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(4), P. 1488 - 1488

Published: Feb. 12, 2024

The term “Liver disease” refers to a broad category of disorders affecting the liver. There are variety common liver ailments, such as hepatitis, cirrhosis, and cancer. Accurate early diagnosis is an emergent demand for prediction disease. Conventional diagnostic techniques, radiological, CT scan, function tests, often time-consuming prone inaccuracies in several cases. An application machine learning (ML) deep (DL) techniques efficient approach diagnosing diseases wide range medical fields. This type machine-related can handle various tasks, image recognition, analysis, classification, because it helps train large datasets learns identify patterns that might not be perceived by humans. paper presented here with evaluation performance DL models on estimation subtyping ailment prognosis. In this manuscript, we propose novel approach, termed CNN+LSTM, which integration convolutional neural network (CNN) long short-term memory (LSTM) networks. results study prove ML used improve prognosis CNN+LSTM model achieves better accuracy 98.73% compared other CNN, Recurrent Neural Network (RNN), LSTM. incorporation proposed has terms (98.73%), precision (99%), recall (98%), F1 score AUC (Area Under Curve)-ROC (Receiver Operating Characteristic) respectively. use shows robustness predicting accurate

Language: Английский

---

ResTransUnet: An effective network combined with Transformer and U-Net for liver segmentation in CT scans

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 177, P. 108625 - 108625

Published: May 21, 2024

Language: Английский

---

Diagnosis of liver disorder

Elsevier eBooks, Journal Year: 2025, Volume and Issue: unknown, P. 199 - 224

Published: Jan. 1, 2025

---