Cascade and Fusion of Multitask Convolutional Neural Networks for Detection of Thyroid Nodules in Contrast-Enhanced CT

Zhen, Zhao; Ye, Chen; Hu, Yanjun; Li, Ceng; Li, Xiaofeng

doi:10.1155/2019/7401235

Cited by 22 publications

(9 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…58 patients had accurate location of the lesion with an accuracy rate of 96.67%, which confirmed that the MSCT scan could clearly show the morphology and structure of the lesion, thereby providing intuitive clinical diagnosis. This was consistent with the research findings of Zhao et al [ 20 ].…”

Section: Discussionsupporting

confidence: 94%

Deep Learning-Based Computed Tomography Image Features in the Detection and Diagnosis of Perianal Abscess Tissue

Han

2021

Journal of Healthcare Engineering

View full text Add to dashboard Cite

The performance characteristics of deep learning fully convolutional neural network (DLFCNN) algorithm-based computed tomography (CT) images were investigated in the detection and diagnosis of perianal abscess tissue. 60 patients who were medically diagnosed as perianal abscesses in the hospital were selected as the experimental group, and 60 healthy volunteers were selected as the control group. In this study, the DLFCNN algorithm based on deep learning was compared with the CNN algorithm and applied to the segmentation training of CT images of patients with perianal abscesses. Then, the segmentation metrics Jaccard, Dice coefficient, precision rate, and recall rate were compared by extracting the region of interest. The results showed that Jaccard (0.7326) calculated by the CNN algorithm was sharply lower than that of the DLFCNN algorithm (0.8525), and the Dice coefficient (0.7264) was also steeply lower than that of the DLFCNN algorithm (0.8434) ( P < 0.05 ). The thickness range of the epidermis and dermis in patients from the experimental group was 4.1–4.9 mm, which was markedly greater than the range of the control group (1.8–3.6 mm) ( P < 0.05 ). Besides, the CT value of the subcutaneous fascia in the experimental group (−95.45 ± 8.26) hugely reduced compared with the control group (−76.34 ± 7.69) ( P < 0.05 ). The accuracy rate of the patients with perianal abscesses was 96.67% by multislice spiral CT (MSCT). Therefore, the DLFCNN algorithm in this study had good stability and good segmentation effect. The skin at the focal site of anal abscess was obviously thickened, and it was simple and accurate to use CT images in the diagnosis of patients with perianal abscesses, which could effectively locate the lesion and clarify the relationship between the lesion and the surrounding structure.

show abstract

Section: Discussionsupporting

confidence: 94%

Deep Learning-Based Computed Tomography Image Features in the Detection and Diagnosis of Perianal Abscess Tissue

Han

2021

Journal of Healthcare Engineering

View full text Add to dashboard Cite

show abstract

“…The recall rate is the ratio of the number of samples that are correctly predicted for the class to the total number of samples; it is also called the sensitivity or hit rate. Precision refers to the ratio of the number of category samples correctly predicted to the total number of samples predicted for that category [ 42 ]. The calculation methods for these indicators are as follows:

where a is the number of samples misclassified, m is the total number of samples, TP is the number of positive samples correctly classified by the model, and FP is the number of negative samples misclassified by the model as positive samples.…”

Section: Experimental Results and Analysismentioning

confidence: 99%

SAI‐YOLO: A Lightweight Network for Real‐Time Detection of Driver Mask‐Wearing Specification on Resource‐Constrained Devices

Zhen

Hao

et al. 2021

Computational Intelligence and Neuroscience

Self Cite

View full text Add to dashboard Cite

Frequent occurrence and long-term existence of respiratory diseases such as COVID-19 and influenza require bus drivers to wear masks correctly during driving. To quickly detect whether the mask is worn correctly on resource-constrained devices, a lightweight target detection network SAI-YOLO is proposed. Based on YOLOv4-Tiny, the network incorporates the Inception V3 structure, replaces two CSPBlock modules with the RES-SEBlock modules to reduce the number of parameters and computational difficulty, and adds a convolutional block attention module and a squeeze-and-excitation module to extract key feature information. Moreover, a modified ReLU (M-ReLU) activation function is introduced to replace the original Leaky_ReLU function. The experimental results show that SAI-YOLO reduces the number of network parameters and calculation difficulty and improves the detection speed of the network while maintaining certain recognition accuracy. The mean average precision (mAP) for face-mask-wearing detection reaches 86% and the average precision (AP) for mask-wearing normative detection reaches 88%. In the resource-constrained device Raspberry Pi 4B, the average detection time after acceleration is 197 ms, which meets the actual application requirements.

show abstract

“…The work conducted by [83] adopted CT scans through CNN for detecting thyroid cancer metastasis, and their work reached an accuracy score of 0.904. Moreover, [84] fused two CNNs for CT scans to detect malignant thyroid nodules and obtained an accuracy of 95.73. While in this study, we obtained an accuracy of 0.966 for the hospital left-side CT scans and 0.970 for the right-side CT scans through the base model.…”

Section: Ctmentioning

confidence: 99%

Multi-channel convolutional neural network architectures for thyroid cancer detection

et al. 2022

View full text Add to dashboard Cite

Early detection of malignant thyroid nodules leading to patient-specific treatments can reduce morbidity and mortality rates. Currently, thyroid specialists use medical images to diagnose then follow the treatment protocols, which have limitations due to unreliable human false-positive diagnostic rates. With the emergence of deep learning, advances in computer-aided diagnosis techniques have yielded promising earlier detection and prediction accuracy; however, clinicians’ adoption is far lacking. The present study adopts Xception neural network as the base structure and designs a practical framework, which comprises three adaptable multi-channel architectures that were positively evaluated using real-world data sets. The proposed architectures outperform existing statistical and machine learning techniques and reached a diagnostic accuracy rate of 0.989 with ultrasound images and 0.975 with computed tomography scans through the single input dual-channel architecture. Moreover, the patient-specific design was implemented for thyroid cancer detection and has obtained an accuracy of 0.95 for double inputs dual-channel architecture and 0.94 for four-channel architecture. Our evaluation suggests that ultrasound images and computed tomography (CT) scans yield comparable diagnostic results through computer-aided diagnosis applications. With ultrasound images obtained slightly higher results, CT, on the other hand, can achieve the patient-specific diagnostic design. Besides, with the proposed framework, clinicians can select the best fitting architecture when making decisions regarding a thyroid cancer diagnosis. The proposed framework also incorporates interpretable results as evidence, which potentially improves clinicians’ trust and hence their adoption of the computer-aided diagnosis techniques proposed with increased efficiency and accuracy.

show abstract

Cascade and Fusion of Multitask Convolutional Neural Networks for Detection of Thyroid Nodules in Contrast-Enhanced CT

Cited by 22 publications

References 30 publications

Deep Learning-Based Computed Tomography Image Features in the Detection and Diagnosis of Perianal Abscess Tissue

Deep Learning-Based Computed Tomography Image Features in the Detection and Diagnosis of Perianal Abscess Tissue

SAI‐YOLO: A Lightweight Network for Real‐Time Detection of Driver Mask‐Wearing Specification on Resource‐Constrained Devices

Multi-channel convolutional neural network architectures for thyroid cancer detection

Contact Info

Product

Resources

About