Semantic Segmentation of the Malignant Breast Imaging Reporting and Data System Lexicon on Breast Ultrasound Images by Using DeepLab v3+

Shia, Wei‐Chung; Hsu, Fang‐Rong; Dai, Seng-Tong; Guo, Shih-Lin; Chen, Dar Ren

doi:10.3390/s22145352

Cited by 9 publications

(6 citation statements)

References 32 publications

(32 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The segmentation method selected for this study was semantic segmentation, which has been successfully used in various applications, including facial attribute prediction [39] and detection of malignancy in breast ultrasound images [40]. This technique is supported by many MATLAB functions, such as "segnetLayers" and "deeplabv3plusLayers."…”

Section: Networkmentioning

confidence: 99%

See 1 more Smart Citation

Detection of Heartbeat Components Through Doppler Radar Systems Using Semantic Segmentation and Non-Harmonic Analysis

Goto,

Horimoto,

Koyama

et al. 2024

IEEE Access

View full text Add to dashboard Cite

The spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has led to an increased focus on the routine analysis of vital signs such as breathing and pulse rates. Radar technology has proven effective for non-contact, long-term monitoring of these vital signs, with frequency analysis being the default method for processing signals from Doppler radar owing to their inherent noise. However, conventional analysis approaches often struggle to detect weak signals buried within the sidelobes of other signals. Some data analysis techniques for Doppler radar rely on machine learning, but they struggle to generate clear time-frequency diagrams, complicating heartbeat detection. In this study, we employed non-harmonic analysis (NHA) as a frequency analysis method to mitigate sidelobe interference and implemented semantic segmentation for precise heartbeat detection. To validate the proposed approach, we conducted heartbeat detection tests both in stationary, low-noise conditions and in a noisy driving simulation environment. The results indicated that the NHA method successfully analyzed heartbeat harmonics, suggesting its potential for detecting heartbeat components through machine learning. To validate these findings, we determined the detection accuracy by comparing true and false positive rates, allowing us to quantify the detectability of heartbeats under both resting and driving simulation conditions.INDEX TERMS Continuous-wave Doppler radar (CW Doppler radar), driving simulation, harmonic, heartbeat, non-harmonic analysis (NHA).

show abstract

Section: Networkmentioning

confidence: 99%

“…In this study, ResNet18 was employed as the CNN model for extracting image features through the encoder. DeepLabv3+ has been implemented using the "deeplabv3plusLayers" function in MATLAB to detect lychee branches [48] and malignant tumors in breast ultrasound images [40].…”

Section: ) Deeplabv3+mentioning

confidence: 99%

Detection of Heartbeat Components Through Doppler Radar Systems Using Semantic Segmentation and Non-Harmonic Analysis

Goto,

Horimoto,

Koyama

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The masking works for the ground truth has essential rules for good performance of the deep learning model. The mask with specific color should cover the only bone regions according to HU value of the CT images, each fractured fragments should be individually separated as the different color masks [37][38][39][40][41][42][43] . There are 38 classes and each class has its own mask color with label (Background, Patella, Femur, Tibia from 1 to 23, Fibula from 1 to 12).…”

Section: Preparation Of Datamentioning

confidence: 99%

“…Many clinical fields require accurate segmented images from digital imaging and communications in medicine (DICOM) for diagnostics, planning, and simulation [29][30][31][32][33][34][35][36] . For example, a tumor region that is difficult to detect can be segmented from an image 17,[37][38][39] . A 3D vessel model can be reconstructed using a 2D segmented image to establish plans for approach and stent insertion 36 .…”

mentioning

confidence: 99%

Automatic segmentation of inconstant fractured fragments for tibia/fibula from CT images using deep learning

Kim,

Jeon,

Park

et al. 2023

Sci Rep

View full text Add to dashboard Cite

Orthopaedic surgeons need to correctly identify bone fragments using 2D/3D CT images before trauma surgery. Advances in deep learning technology provide good insights into trauma surgery over manual diagnosis. This study demonstrates the application of the DeepLab v3+ -based deep learning model for the automatic segmentation of fragments of the fractured tibia and fibula from CT images and the results of the evaluation of the performance of the automatic segmentation. The deep learning model, which was trained using over 11 million images, showed good performance with a global accuracy of 98.92%, a weighted intersection over the union of 0.9841, and a mean boundary F1 score of 0.8921. Moreover, deep learning performed 5–8 times faster than the experts’ recognition performed manually, which is comparatively inefficient, with almost the same significance. This study will play an important role in preoperative surgical planning for trauma surgery with convenience and speed.

show abstract

“…In recent years, computer vision and its related fields have been developing very rapidly and are playing an increasingly important role in aiding disease diagnosis, including colorectal Cancer [17,18] , thyroid nodules [19,20] , congenital heart disease [21,22] , Alzheimer's disease [23,24] , breast cancer [25][26][27] , dermatology [28][29][30] and screening of abnormality [31][32][33] . Progress has been made in applying deep learning to different medical application tasks, such as image classification [19,34,35] , semantic segmentation [36][37][38] , object detection [39][40][41][42] , instance segmentation [43,44] . Computer Aide Diagnosis (CAD) of NEC is not a new topic.…”

Section: Introductionmentioning

confidence: 99%

A novel two-stage deep learning model used to assist in diagnosing neonatal necrotizing enterocolitis and determining the need for surgical treatment

Qi,

Ding,

et al. 2023

Preprint

View full text Add to dashboard Cite

Background and Aims: Neonatal necrotizing enterocolitis (NEC) is a common life-threatening gastrointestinal disease in newborns. Abdominal X-rays (AXRs) is an important basis for diagnosing NEC and determining the need for surgical treatment. Computer-aided diagnosis (CAD) is extensively utilized in the clinical diagnosis of numerous diseases. Nevertheless, the efficacy of CAD for NEC has not been widely validated. Methods: We proposed for the first time a two-stage multimodal classification method for NEC based on AXRs data. The objective is to achieve early diagnosis of NEC and determine the optimal timing for surgical intervention. This method addresses the problem of insufficient labeled data through transfer learning and introduces coordinate attention to enhance the accuracy of target region localization and identification, thereby improving the capability of image feature extraction. Results: In total, the dataset was sourced from 2 children’s hospital consisted of 3,176 AXRs from 845 newborns diagnosed with NEC. Additionally, there were 1,825 AXRs from 470 newborns without NEC. The task for determining whether newborns has NEC achieved an accuracy of 97.49%, recall of 97.44%, precision of 83.09%, F1-score of 98.02% and AUC of 99.68%. Similarly, for the task of identifying if NEC patients require surgery, the accuracy, recall, precision, and F1-score were 78.96%, 81.50%, 80.30%, 80.89%, and 84.49% respectively. Our method performed better than the four commonly used baseline methods in the two-stage NEC diagnosis task. Conclusions: We have introduced a novel two-stage diagnostic model for NEC in newborns, which can rapidly and accurately identify NEC patients and determine if surgery is necessary.

show abstract

Semantic Segmentation of the Malignant Breast Imaging Reporting and Data System Lexicon on Breast Ultrasound Images by Using DeepLab v3+

Cited by 9 publications

References 32 publications

Detection of Heartbeat Components Through Doppler Radar Systems Using Semantic Segmentation and Non-Harmonic Analysis

Detection of Heartbeat Components Through Doppler Radar Systems Using Semantic Segmentation and Non-Harmonic Analysis

Automatic segmentation of inconstant fractured fragments for tibia/fibula from CT images using deep learning

A novel two-stage deep learning model used to assist in diagnosing neonatal necrotizing enterocolitis and determining the need for surgical treatment

Contact Info

Product

Resources

About