Research

14 pages, 708 KiB

Open AccessArticle

Neural Network Model Based on Branch Architecture for the Quality Assurance of Volumetric Modulated Arc Therapy

by Lizhang Xie, Lei Zhang, Ting Hu, Guangjun Li and Zhang Yi

Bioengineering 2024, 11(4), 362; https://doi.org/10.3390/bioengineering11040362 - 11 Apr 2024

Viewed by 345

Radiation therapy relies on quality assurance (QA) to verify dose delivery accuracy. However, current QA methods suffer from operation lag as well as inaccurate performance. Hence, to address these shortcomings, this paper proposes a QA neural network model based on branch architecture, which [...] Read more.

Radiation therapy relies on quality assurance (QA) to verify dose delivery accuracy. However, current QA methods suffer from operation lag as well as inaccurate performance. Hence, to address these shortcomings, this paper proposes a QA neural network model based on branch architecture, which is based on the analysis of the category features of the QA complexity metrics. The designed branch network focuses on category features, which effectively improves the feature extraction capability for complexity metrics. The branch features extracted by the model are fused to predict the GPR for more accurate QA. The performance of the proposed method was validated on the collected dataset. The experiments show that the prediction performance of the model outperforms other QA methods; the average prediction errors for the test set are 2.12% (2%/2 mm), 1.69% (3%/2 mm), and 1.30% (3%/3 mm). Moreover, the results indicate that two-thirds of the validation samples’ model predictions perform better than the clinical evaluation results, suggesting that the proposed model can assist physicists in the clinic. Full article

(This article belongs to the Special Issue Artificial Intelligence in Auto-Diagnosis and Clinical Applications)

► Show Figures

Figure 1

11 pages, 3728 KiB

Open AccessArticle

SpineHRformer: A Transformer-Based Deep Learning Model for Automatic Spine Deformity Assessment with Prospective Validation

by Moxin Zhao, Nan Meng, Jason Pui Yin Cheung, Chenxi Yu, Pengyu Lu and Teng Zhang

Bioengineering 2023, 10(11), 1333; https://doi.org/10.3390/bioengineering10111333 - 20 Nov 2023

Viewed by 1045

Abstract

The Cobb angle (CA) serves as the principal method for assessing spinal deformity, but manual measurements of the CA are time-consuming and susceptible to inter- and intra-observer variability. While learning-based methods, such as SpineHRNet+, have demonstrated potential in automating CA measurement, their accuracy [...] Read more.

The Cobb angle (CA) serves as the principal method for assessing spinal deformity, but manual measurements of the CA are time-consuming and susceptible to inter- and intra-observer variability. While learning-based methods, such as SpineHRNet+, have demonstrated potential in automating CA measurement, their accuracy can be influenced by the severity of spinal deformity, image quality, relative position of rib and vertebrae, etc. Our aim is to create a reliable learning-based approach that provides consistent and highly accurate measurements of the CA from posteroanterior (PA) X-rays, surpassing the state-of-the-art method. To accomplish this, we introduce SpineHRformer, which identifies anatomical landmarks, including the vertices of endplates from the 7th cervical vertebra (C7) to the 5th lumbar vertebra (L5) and the end vertebrae with different output heads, enabling the calculation of CAs. Within our SpineHRformer, a backbone HRNet first extracts multi-scale features from the input X-ray, while transformer blocks extract local and global features from the HRNet outputs. Subsequently, an output head to generate heatmaps of the endplate landmarks or end vertebra landmarks facilitates the computation of CAs. We used a dataset of 1934 PA X-rays with diverse degrees of spinal deformity and image quality, following an 8:2 ratio to train and test the model. The experimental results indicate that SpineHRformer outperforms SpineHRNet+ in landmark detection (Mean Euclidean Distance: 2.47 pixels vs. 2.74 pixels), CA prediction (Pearson correlation coefficient: 0.86 vs. 0.83), and severity grading (sensitivity: normal-mild; 0.93 vs. 0.74, moderate; 0.74 vs. 0.77, severe; 0.74 vs. 0.7). Our approach demonstrates greater robustness and accuracy compared to SpineHRNet+, offering substantial potential for improving the efficiency and reliability of CA measurements in clinical settings. Full article

(This article belongs to the Special Issue Artificial Intelligence in Auto-Diagnosis and Clinical Applications)

► Show Figures

Figure 1

15 pages, 3645 KiB

Open AccessArticle

Automatic Segmentation and Assessment of Valvular Regurgitations with Color Doppler Echocardiography Images: A VABC-UNet-Based Framework

by Jun Huang, Aiyue Huang, Ruqin Xu, Musheng Wu, Peng Wang and Qing Wang

Bioengineering 2023, 10(11), 1319; https://doi.org/10.3390/bioengineering10111319 - 16 Nov 2023

Viewed by 898

Abstract

This study investigated the automatic segmentation and classification of mitral regurgitation (MR) and tricuspid regurgitation (TR) using a deep learning-based method, aiming to improve the efficiency and accuracy of diagnosis of valvular regurgitations. A VABC-UNet model was proposed consisting of VGG16 encoder, U-Net [...] Read more.

This study investigated the automatic segmentation and classification of mitral regurgitation (MR) and tricuspid regurgitation (TR) using a deep learning-based method, aiming to improve the efficiency and accuracy of diagnosis of valvular regurgitations. A VABC-UNet model was proposed consisting of VGG16 encoder, U-Net decoder, batch normalization, attention block and deepened convolution layer based on the U-Net backbone. Then, a VABC-UNet-based assessment framework was established for automatic segmentation, classification, and evaluation of valvular regurgitations. A total of 315 color Doppler echocardiography images of MR and/or TR in an apical four-chamber view were collected, including 35 images in the test dataset and 280 images in the training dataset. In comparison with the classic U-Net and VGG16-UNet models, the segmentation performance of the VABC-UNet model was evaluated via four metrics: Dice, Jaccard, Precision, and Recall. According to the features of regurgitation jet and atrium, the regurgitation could automatically be classified into MR or TR, and evaluated to mild, moderate, moderate–severe, or severe grade by the framework. The results show that the VABC-UNet model has a superior performance in the segmentation of valvular regurgitation jets and atria to the other two models and consequently a higher accuracy of classification and evaluation. There were fewer pseudo- and over-segmentations by the VABC-UNet model and the values of the metrics significantly improved (p < 0.05). The proposed VABC-UNet-based framework achieves automatic segmentation, classification, and evaluation of MR and TR, having potential to assist radiologists in clinical decision making of the regurgitations in valvular heart diseases. Full article

(This article belongs to the Special Issue Artificial Intelligence in Auto-Diagnosis and Clinical Applications)

► Show Figures

Graphical abstract

24 pages, 1634 KiB

Open AccessArticle

RCKD: Response-Based Cross-Task Knowledge Distillation for Pathological Image Analysis

by Hyunil Kim, Tae-Yeong Kwak, Hyeyoon Chang, Sun Woo Kim and Injung Kim

Bioengineering 2023, 10(11), 1279; https://doi.org/10.3390/bioengineering10111279 - 02 Nov 2023

Viewed by 1112

Abstract

We propose a novel transfer learning framework for pathological image analysis, the Response-based Cross-task Knowledge Distillation (RCKD), which improves the performance of the model by pretraining it on a large unlabeled dataset guided by a high-performance teacher model. RCKD first pretrains a student [...] Read more.

We propose a novel transfer learning framework for pathological image analysis, the Response-based Cross-task Knowledge Distillation (RCKD), which improves the performance of the model by pretraining it on a large unlabeled dataset guided by a high-performance teacher model. RCKD first pretrains a student model to predict the nuclei segmentation results of the teacher model for unlabeled pathological images, and then fine-tunes the pretrained model for the downstream tasks, such as organ cancer sub-type classification and cancer region segmentation, using relatively small target datasets. Unlike conventional knowledge distillation, RCKD does not require that the target tasks of the teacher and student models be the same. Moreover, unlike conventional transfer learning, RCKD can transfer knowledge between models with different architectures. In addition, we propose a lightweight architecture, the Convolutional neural network with Spatial Attention by Transformers (CSAT), for processing high-resolution pathological images with limited memory and computation. CSAT exhibited a top-1 accuracy of 78.6% on ImageNet with only 3M parameters and 1.08 G multiply-accumulate (MAC) operations. When pretrained by RCKD, CSAT exhibited average classification and segmentation accuracies of 94.2% and 0.673 mIoU on six pathological image datasets, which is 4% and 0.043 mIoU higher than EfficientNet-B0, and 7.4% and 0.006 mIoU higher than ConvNextV2-Atto pretrained on ImageNet, respectively. Full article

(This article belongs to the Special Issue Artificial Intelligence in Auto-Diagnosis and Clinical Applications)

► Show Figures

Figure 1

15 pages, 2637 KiB

Open AccessArticle

Hybrid Machine Learning Framework for Multistage Parkinson’s Disease Classification Using Acoustic Features of Sustained Korean Vowels

by S. I. M. M. Raton Mondol, Ryul Kim and Sangmin Lee

Bioengineering 2023, 10(8), 984; https://doi.org/10.3390/bioengineering10080984 - 20 Aug 2023

Viewed by 1198

Abstract

Recent research has achieved a great classification rate for separating healthy people from those with Parkinson’s disease (PD) using speech and the voice. However, these studies have primarily treated early and advanced stages of PD as equal entities, neglecting the distinctive speech impairments [...] Read more.

Recent research has achieved a great classification rate for separating healthy people from those with Parkinson’s disease (PD) using speech and the voice. However, these studies have primarily treated early and advanced stages of PD as equal entities, neglecting the distinctive speech impairments and other symptoms that vary across the different stages of the disease. To address this limitation, and improve diagnostic precision, this study assesses the selected acoustic features of dysphonia, as they relate to PD and the Hoehn and Yahr stages, by combining various preprocessing techniques and multiple classification algorithms, to create a comprehensive and robust solution for classification tasks. The dysphonia features extracted from the three sustained Korean vowels /아/(a), /이/(i), and /우/(u) exhibit diversity and strong correlations. To address this issue, the analysis of variance F-Value feature selection classifier from scikit-learn was employed, to identify the topmost relevant features. Additionally, to overcome the class imbalance problem, the synthetic minority over-sampling technique was utilized. To ensure fair comparisons, and mitigate the influence of individual classifiers, four commonly used machine learning classifiers, namely random forest (RF), support vector machine (SVM), k-nearest neighbor (kNN), and multi-layer perceptron (MLP), were employed. This approach enables a comprehensive evaluation of the feature extraction methods, and minimizes the variance in the final classification models. The proposed hybrid machine learning pipeline using the acoustic features of sustained vowels efficiently detects the early and mid-advanced stages of PD with a detection accuracy of 95.48%, and with a detection accuracy of 86.62% for the 4-stage, and a detection accuracy of 89.48% for the 3-stage classification of PD. This study successfully demonstrates the significance of utilizing the diverse acoustic features of dysphonia in the classification of PD and its stages. Full article

(This article belongs to the Special Issue Artificial Intelligence in Auto-Diagnosis and Clinical Applications)

► Show Figures

Figure 1

14 pages, 2233 KiB

Open AccessArticle

Effect of Cilostazol on Delayed Cerebral Infarction in Aneurysmal Subarachnoid Hemorrhage Using Explainable Predictive Modeling

by Kwang Hyeon Kim, Byung-Jou Lee and Hae-Won Koo

Bioengineering 2023, 10(7), 797; https://doi.org/10.3390/bioengineering10070797 - 03 Jul 2023

Cited by 1 | Viewed by 1207

Abstract

The studies interpreting DCI, a complication of SAH, and identifying correlations are very limited. This study aimed to investigate the effect of cilostazol on ACV and DCI after coil embolization for ruptured aneurysms (n = 432). A multivariate analysis was performed and explainable [...] Read more.

The studies interpreting DCI, a complication of SAH, and identifying correlations are very limited. This study aimed to investigate the effect of cilostazol on ACV and DCI after coil embolization for ruptured aneurysms (n = 432). A multivariate analysis was performed and explainable artificial intelligence approaches were used to analyze the contribution of cilostazol as a risk factor on the development of ACV and DCI with respect to global and local interpretation. The cilonimo group was significantly lower than the nimo group in ACV (13.5% vs. 29.3; p = 0.003) and DCI (7.9% vs. 20.7%; p = 0.006), respectively. In a multivariate logistic regression, the odds ratio for DCI for the cilonimo group, female sex, and aneurysm size was 0.556 (95% confidence interval (CI), 0.351–0.879; p = 0.012), 3.713 (95% CI, 1.683–8.191; p = 0.001), and 1.106 (95% CI, 1.008–1.214; p = 0.034). The risk of a DCI occurrence was significantly increased with an aneurysm size greater than 10 mm (max 80%). The mean AUC of the XGBoost and logistic regression models was 0.94 ± 0.03 and 0.95 ± 0.04, respectively. Cilostazol treatment combined with nimodipine could decrease the prevalence of ACV (13.5%) and DCI (7.9%) in patients with aSAH. Full article

(This article belongs to the Special Issue Artificial Intelligence in Auto-Diagnosis and Clinical Applications)

► Show Figures

Figure 1

12 pages, 2131 KiB

Open AccessArticle

Identification of Adolescent Menarche Status Using Biplanar X-ray Images: A Deep Learning-Based Method

by Linzhen Xie, Tenghui Ge, Bin Xiao, Xiaoguang Han, Qi Zhang, Zhongning Xu, Da He and Wei Tian

Bioengineering 2023, 10(7), 769; https://doi.org/10.3390/bioengineering10070769 - 26 Jun 2023

Viewed by 827

Abstract

The purpose of this study is to develop an automated method for identifying the menarche status of adolescents based on EOS radiographs. We designed a deep-learning-based algorithm that contains a region of interest detection network and a classification network. The algorithm was trained [...] Read more.

The purpose of this study is to develop an automated method for identifying the menarche status of adolescents based on EOS radiographs. We designed a deep-learning-based algorithm that contains a region of interest detection network and a classification network. The algorithm was trained and tested on a retrospective dataset of 738 adolescent EOS cases using a five-fold cross-validation strategy and was subsequently tested on a clinical validation set of 259 adolescent EOS cases. On the clinical validation set, our algorithm achieved accuracy of 0.942, macro precision of 0.933, macro recall of 0.938, and a macro F1-score of 0.935. The algorithm showed almost perfect performance in distinguishing between males and females, with the main classification errors found in females aged 12 to 14 years. Specifically for females, the algorithm had accuracy of 0.910, sensitivity of 0.943, and specificity of 0.855 in estimating menarche status, with an area under the curve of 0.959. The kappa value of the algorithm, in comparison to the actual situation, was 0.806, indicating strong agreement between the algorithm and the real-world scenario. This method can efficiently analyze EOS radiographs and identify the menarche status of adolescents. It is expected to become a routine clinical tool and provide references for doctors’ decisions under specific clinical conditions. Full article

(This article belongs to the Special Issue Artificial Intelligence in Auto-Diagnosis and Clinical Applications)

► Show Figures

Graphical abstract

21 pages, 5493 KiB

Open AccessArticle

LcmUNet: A Lightweight Network Combining CNN and MLP for Real-Time Medical Image Segmentation

by Shuai Zhang and Yanmin Niu

Bioengineering 2023, 10(6), 712; https://doi.org/10.3390/bioengineering10060712 - 12 Jun 2023

Cited by 2 | Viewed by 1995

Abstract

In recent years, UNet and its improved variants have become the main methods for medical image segmentation. Although these models have achieved excellent results in segmentation accuracy, their large number of network parameters and high computational complexity make it difficult to achieve medical [...] Read more.

In recent years, UNet and its improved variants have become the main methods for medical image segmentation. Although these models have achieved excellent results in segmentation accuracy, their large number of network parameters and high computational complexity make it difficult to achieve medical image segmentation in real-time therapy and diagnosis rapidly. To address this problem, we introduce a lightweight medical image segmentation network (LcmUNet) based on CNN and MLP. We designed LcmUNet’s structure in terms of model performance, parameters, and computational complexity. The first three layers are convolutional layers, and the last two layers are MLP layers. In the convolution part, we propose an LDA module that combines asymmetric convolution, depth-wise separable convolution, and an attention mechanism to reduce the number of network parameters while maintaining a strong feature-extraction capability. In the MLP part, we propose an LMLP module that helps enhance contextual information while focusing on local information and improves segmentation accuracy while maintaining high inference speed. This network also covers skip connections between the encoder and decoder at various levels. Our network achieves real-time segmentation results accurately in extensive experiments. With only 1.49 million model parameters and without pre-training, LcmUNet demonstrated impressive performance on different datasets. On the ISIC2018 dataset, it achieved an IoU of 85.19%, 92.07% recall, and 92.99% precision. On the BUSI dataset, it achieved an IoU of 63.99%, 79.96% recall, and 76.69% precision. Lastly, on the Kvasir-SEG dataset, LcmUNet achieved an IoU of 81.89%, 88.93% recall, and 91.79% precision. Full article

(This article belongs to the Special Issue Artificial Intelligence in Auto-Diagnosis and Clinical Applications)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Artificial Intelligence in Auto-Diagnosis and Clinical Applications

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Published Papers (8 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI