Application of Deep Learning as a Noninvasive Tool to Determine Pathological Diagnosis of Enlarged Cervical Lymph Nodes with PET/CT

2020 ◽  
Author(s):  
Yuhan Yang ◽  
Bo Zheng ◽  
Yixi Wang ◽  
Xuelei Ma
Keyword(s):  
Deep Learning ◽  
Lymph Nodes ◽  
Diagnostic Performance ◽  
Imaging Modality ◽  
Ct Images ◽  
Pathological Diagnosis ◽  
Therapeutic Interventions ◽  
Cross Sectional ◽  
Cervical Lymph Nodes ◽  
Pet Ct

Abstract Objective: To construct a deep-learning convolution neural network (DL-CNN) system for pathological diagnosis of cervical lymph nodes by using computed tomography (CT), fluorine-18 fluorodeoxyglucose (18F-FDG) positron emission tomography (PET), and fused PET/CT images.Materials and methods: A total of 1020 cross-sectional images for each imaging modality was obtained from 211 patients (153 patients with lymphomas and 116 patients with metastases) with enlarged cervical lymph nodes from January 2014 to June 2018. All eligible images were distributed randomly into the training, validation, and testing cohorts with ratios of 70%, 15%, and 15%. We applied eight DL-CNN algorithms with pretrained bases from ImageNet dataset on CT, PET, and fused PET/CT imaging datasets to differentiate lymphomatous nodes from metastatic nodes, respectively. Attention heatmaps of PET and fused PET/CT images generated by class activation mapping (CAM) were used in visualization of class specific regions recognized by the prediction model with best performance. Results: The accuracy of eight deep learning algorithms with pretrained base ranged from 0.650 to 0.981 on PET testing cohort, and from 0.738 to 0.981 on fused PET/CT testing cohort. The VGG16 model on PET images and DenseNet121 model on fused PET/CT images had the best diagnostic performance among all eight algorithms with sensitivity and specificity of 1.000 and 0.963. Class-specific discriminative subregions were highlighted by attention maps for clinical review.Conclusion: A DL-CNN system was developed for classifying metastatic and lymphomatous involvement with favorable diagnostic performance on PET and PET/CT images in patients with enlarged cervical lymph nodes. The further clinical practice of this system may improve quality of the following therapeutic interventions and optimize patients’ outcomes.

scholarly journals Brain metastases assessment by FDG-PET/CT: can it eliminate the necessity for dedicated brain imaging?

2020 ◽  
Vol 51 (1) ◽  
Author(s):  
Mamdouh A. Zidan ◽  
Radwa S. Hassan ◽  
Khaled I. El-Noueam ◽  
Yasser M. Zakaria
Keyword(s):  
Brain Metastases ◽  
Fdg Pet ◽  
Diagnostic Performance ◽  
Imaging Modality ◽  
False Negative ◽  
Whole Body ◽  
Cross Sectional ◽  
Primary Tumors ◽  
Pet Ct ◽  
Fdg Pet Ct

Abstract Background Brain metastases (BM) are the most common intracranial tumors in adults outnumbering all other intracranial neoplasms. Positron emission tomography combined with computed tomography (PET/CT) is a widely used imaging modality in oncology with a unique combination of cross-sectional anatomic information provided by CT and the metabolic information provided by PET using the [18F]-2-fluoro-2-deoxy-d-glucose (FDG) as a tracer. The aim of the study is to assess the role and diagnostic performance of brain-included whole-body PET/CT in detection and evaluation of BM and when further imaging is considered necessary. The study was conducted over a period of 12 months on 420 patients suffering from extra-cranial malignancies utilizing brain-included whole-body PET/CT. Results Thirty patients with 71 brain lesions were detected, 18 patients (60%) had BM of unknown origin while 12 patients (40%) presented with known primary tumors. After brain-included whole-body FDG-PET/CT examination, the unknown primaries turned out to be bronchogenic carcinoma in 10 patients (33.3%), renal cell carcinoma in 2 patients (6.7%), and lymphoma in 2 patients (6.7%), yet the primary tumors remained unknown in 4 patients (13.3%). In 61 lesions (85.9%), the max SUV ranged from 0.2- < 10, while in 10 lesions (14.1%) the max SUV ranged from 10 to 20. Hypometabolic lesions were reported in 41 (57.7%) lesions, hypermetabolic in 3 lesions (4.2%), whereas 27 lesions (38.0%) showed similar FDG uptake to the corresponding contralateral brain matter. PET/CT overall sensitivity, specificity, positive and negative predictive, and accuracy values were 78.1, 92.6, 83.3, 90, and 88% respectively. Conclusion Brain-included whole-body FDG-PET/CT provides valuable complementary information in the evaluation of patients with suspected BM. However, the diagnostic performance of brain PET-CT carries the possibility of false-negative results with consequent false sense of security. The clinicians should learn about the possible pitfalls of PET/CT interpretation to direct patients with persistent neurological symptoms or high suspicion for BM for further dedicated CNS imaging.

scholarly journals Cervical Lymph Nodes Detected by F-18 FDG PET/CT in Oncology Patients: Added Value of Subsequent Ultrasonography for Determining Nodal Metastasis

Medicina ◽  
2019 ◽  
Vol 56 (1) ◽  
pp. 16
Author(s):  
Seokho Yoon ◽  
Kyeong Hwa Ryu ◽  
Hye Jin Baek ◽  
Tae Hoon Kim ◽  
Jin Il Moon ◽  
...  
Keyword(s):  
Lymph Nodes ◽  
Diagnostic Performance ◽  
Standardized Uptake Value ◽  
Nodal Metastasis ◽  
Added Value ◽  
Diagnostic Strategy ◽  
Oncology Patients ◽  
Cervical Lymph Nodes ◽  
Pet Ct ◽  
Significant Difference

Background and Objectives: To investigate the diagnostic performance of F-18 fluorodeoxyglucose positron emission tomography/computed tomography (PET/CT) and subsequent ultrasonography (US) for determining cervical nodal metastasis in oncology patients. Materials and Methods: Fifty-nine cervical lymph nodes (LNs) initially detected by PET/CT with subsequent neck US were included in this retrospective study. All LNs were subjected to US-guided fine-needle aspiration or core needle biopsy. The maximum standardized uptake value (SUVmax) and sonographic features were assessed. Results: Forty-three of 59 cervical LNs detected by PET/CT were malignant. PET/CT alone showed a highest diagnostic value for metastatic LNs with 81.4% sensitivity, 68.8% specificity, and 78% accuracy when SUVmax ≥5.8 was applied as an optimal cut-off value. Combined PET/CT and subsequent US diagnoses for determining nodal metastasis showed the following diagnostic performance: 81.4% sensitivity, 87.5% specificity, and 83.1% accuracy. There was a significant difference in the diagnostic performance between the two diagnostic imaging approaches (p = 0.006). Conclusions: Combined diagnosis using subsequent US showed a significantly higher diagnostic performance for determining nodal metastasis in the neck. Therefore, we believe that our proposed diagnostic strategy using subsequent US can be helpful in evaluating cervical LNs on PET/CT. Moreover, our results clarify the need for US-guided tissue sampling in oncology patients.

Computer aided diagnosis system for cervical lymph nodes in CT images using deep learning

2022 ◽  
Vol 71 ◽  
pp. 103158
Author(s):  
Hitesh Tekchandani ◽  
Shrish Verma ◽  
Narendra D. Londhe ◽  
Rajiv Ratan Jain ◽  
Avani Tiwari
Keyword(s):  
Deep Learning ◽  
Lymph Nodes ◽  
Ct Images ◽  
Computer Aided Diagnosis ◽  
Cervical Lymph Nodes ◽  
Diagnosis System ◽  
Computer Aided ◽  
Aided Diagnosis

COVID-19 pneumonia diagnosis using a simple 2D deep learning framework with a single chest CT image (Preprint)

2020 ◽  
Author(s):  
Jinseok Lee
Keyword(s):  
Deep Learning ◽  
Diagnostic Performance ◽  
Ct Images ◽  
Chest Ct ◽  
University Hospital ◽  
Detection Accuracy ◽  
Ct Image ◽  
Test Dataset ◽  
Learning Framework ◽  
Testing Dataset

BACKGROUND The coronavirus disease (COVID-19) has explosively spread worldwide since the beginning of 2020. According to a multinational consensus statement from the Fleischner Society, computed tomography (CT) can be used as a relevant screening tool owing to its higher sensitivity for detecting early pneumonic changes. However, physicians are extremely busy fighting COVID-19 in this era of worldwide crisis. Thus, it is crucial to accelerate the development of an artificial intelligence (AI) diagnostic tool to support physicians. OBJECTIVE We aimed to quickly develop an AI technique to diagnose COVID-19 pneumonia and differentiate it from non-COVID pneumonia and non-pneumonia diseases on CT. METHODS A simple 2D deep learning framework, named fast-track COVID-19 classification network (FCONet), was developed to diagnose COVID-19 pneumonia based on a single chest CT image. FCONet was developed by transfer learning, using one of the four state-of-art pre-trained deep learning models (VGG16, ResNet50, InceptionV3, or Xception) as a backbone. For training and testing of FCONet, we collected 3,993 chest CT images of patients with COVID-19 pneumonia, other pneumonia, and non-pneumonia diseases from Wonkwang University Hospital, Chonnam National University Hospital, and the Italian Society of Medical and Interventional Radiology public database. These CT images were split into a training and a testing set at a ratio of 8:2. For the test dataset, the diagnostic performance to diagnose COVID-19 pneumonia was compared among the four pre-trained FCONet models. In addition, we tested the FCONet models on an additional external testing dataset extracted from the embedded low-quality chest CT images of COVID-19 pneumonia in recently published papers. RESULTS Of the four pre-trained models of FCONet, the ResNet50 showed excellent diagnostic performance (sensitivity 99.58%, specificity 100%, and accuracy 99.87%) and outperformed the other three pre-trained models in testing dataset. In additional external test dataset using low-quality CT images, the detection accuracy of the ResNet50 model was the highest (96.97%), followed by Xception, InceptionV3, and VGG16 (90.71%, 89.38%, and 87.12%, respectively). CONCLUSIONS The FCONet, a simple 2D deep learning framework based on a single chest CT image, provides excellent diagnostic performance in detecting COVID-19 pneumonia. Based on our testing dataset, the ResNet50-based FCONet might be the best model, as it outperformed other FCONet models based on VGG16, Xception, and InceptionV3.

scholarly journals Ossifying fibrous epulis as an IgG4-related disease of the oral cavity: a case report and literature review

2022 ◽  
Vol 22 (1) ◽  
Author(s):  
Yoshiko Ike ◽  
Takahiro Shimizu ◽  
Masaru Ogawa ◽  
Takahiro Yamaguchi ◽  
Keisuke Suzuki ◽  
...  
Keyword(s):  
Oral Cavity ◽  
Lymph Nodes ◽  
Plasma Cells ◽  
Maxillofacial Surgery ◽  
The Body ◽  
Pathological Diagnosis ◽  
Cervical Lymph Nodes ◽  
Related Disease ◽  
Igg4 Related Disease ◽  
Organ Manifestations

Abstract Background Fibrous sclerosing tumours and hypertrophic lesions in IgG4-related disease (IgG4-RD) are formed in various organs throughout the body, but disease in the oral region is not included among individual organ manifestations. We report a case of ossifying fibrous epulis that developed from the gingiva, as an instance of IgG4-RD. Case presentation A 60-year-old Japanese man visited the Department of Oral and Maxillofacial Surgery, Gunma University Hospital, with a chief complaint of swelling of the left mandibular gingiva. A 65 mm × 45 mm pedunculated tumour was observed. The bilateral submandibular lymph nodes were enlarged. The intraoperative pathological diagnosis of the enlarged cervical lymph nodes was inflammation. Based on this diagnosis, surgical excision was limited to the intraoral tumour, which was subsequently pathologically diagnosed as ossifying fibrous epulis. Histopathologically, the ossifying fibrous epulis exhibited increased levels of fibroblasts and collagen fibres, as well as infiltration by numerous plasma cells. The IgG4/IgG cell ratio was > 40%. Serologic analysis revealed hyper-IgG4-emia (> 135 mg/dL). The patient met the comprehensive clinical diagnosis criteria and the American College of Rheumatology and European League Against Rheumatism classification criteria for IgG4-RD. Based on these criteria, we diagnosed the ossifying fibrous epulis in our patient as an IgG4-related disease. A pathological diagnosis of IgG4-related lymphadenopathy was established for the cervical lymph nodes. Concomitant clinical findings were consistent with type II IgG4-related lymphadenopathy. Conclusions A routine serological test may be needed in cases with marked fibrous changes (such as epulis) in the oral cavity and plasma cells, accompanied by tumour formation, to determine the possibility of individual-organ manifestations of IgG4-related disease.

Assessment of therapeutic response of lymphoma patients by PET/CT

QJM ◽  
2021 ◽  
Vol 114 (Supplement_1) ◽  
Author(s):  
Aya Y Ahmed ◽  
Mona A Nagi ◽  
Radwa H. El Sheikh
Keyword(s):  
Therapeutic Response ◽  
Imaging Modality ◽  
Treatment Plan ◽  
Response Assessment ◽  
Response To Therapy ◽  
University Hospital ◽  
Lymphoid Tissues ◽  
Cross Sectional ◽  
End Of Treatment ◽  
Pet Ct

Abstract Background Lymphoma compromises а histologically heterogeneous group of cancers derived from the cells of the immune system. The hallmark of the disease is the enlargement and proliferation of lymph nodes or secondary lymphoid tissues. Aim of the Work to evaluate the role of Positron emission tomography in the assessment of response to therapy in lymphoma patients: in particular, a five-point scale (Deauville criteria), which can be employed for early- and late-therapeutic response assessment. Patients and Methods This cross sectional study was conducted on 20 Patients with different types of lymphoma recruited and enrolled from Ain Shams university hospital. Results PET/CT and Contrast enhanced computed tomography were concurrent in results in 55% of cases during treatment and 75 % at the end of treatment with CT sensitivity of 61.1%, specificity of 92.2% and accuracy of 76.2% during treatment in comparison to 100 % sensitivity and specificity of PET/CT.While sensitivity of CT at end of treatment is 57.5% with specificity of 86.7% and accuracy of 71.6%. Conclusion PET/СT using 2-deoxy-2[18F]fluoro-D-glucose is considered one of the best oncologic imaging modality at the time being with valuable applications in lymphoma.It is very efficient with least possible pitfalls and false results compared to either of its components alone and to side by side reading of separately acquired PET and СT. It is becoming а standard modality for lymphoma providing а new vision to management and treatment plan.

scholarly journals Reducing the number of unnecessary biopsies of US-BI-RADS 4a lesions through a deep learning method for residents-in-training: a cross-sectional study

BMJ Open ◽  
2020 ◽  
Vol 10 (6) ◽  
pp. e035757
Author(s):  
Chenyang Zhao ◽  
Mengsu Xiao ◽  
He Liu ◽  
Ming Wang ◽  
Hongyan Wang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Breast Imaging ◽  
Diagnostic Performance ◽  
Intraclass Correlation ◽  
Cross Sectional Study ◽  
Computer Assisted ◽  
Sectional Study ◽  
Benign Lesions ◽  
Cross Sectional ◽  
Experienced Radiologist

ObjectiveThe aim of the study is to explore the potential value of S-Detect for residents-in-training, a computer-assisted diagnosis system based on deep learning (DL) algorithm.MethodsThe study was designed as a cross-sectional study. Routine breast ultrasound examinations were conducted by an experienced radiologist. The ultrasonic images of the lesions were retrospectively assessed by five residents-in-training according to the Breast Imaging Report and Data System (BI-RADS) lexicon, and a dichotomic classification of the lesions was provided by S-Detect. The diagnostic performances of S-Detect and the five residents were measured and compared using the pathological results as the gold standard. The category 4a lesions assessed by the residents were downgraded to possibly benign as classified by S-Detect. The diagnostic performance of the integrated results was compared with the original results of the residents.ParticipantsA total of 195 focal breast lesions were consecutively enrolled, including 82 malignant lesions and 113 benign lesions.ResultsS-Detect presented higher specificity (77.88%) and area under the curve (AUC) (0.82) than the residents (specificity: 19.47%–48.67%, AUC: 0.62–0.74). A total of 24, 31, 38, 32 and 42 identified as BI-RADS 4a lesions by residents 1, 2, 3, 4 and 5 were downgraded to possibly benign lesions by S-Detect, respectively. Among these downgraded lesions, 24, 28, 35, 30 and 40 lesions were proven to be pathologically benign, respectively. After combining the residents' results with the results of the software in category 4a lesions, the specificity and AUC of the five residents significantly improved (specificity: 46.02%–76.11%, AUC: 0.71–0.85, p<0.001). The intraclass correlation coefficient of the five residents also increased after integration (from 0.480 to 0.643).ConclusionsWith the help of the DL software, the specificity, overall diagnostic performance and interobserver agreement of the residents greatly improved. The software can be used as adjunctive tool for residents-in-training, downgrading 4a lesions to possibly benign and reducing unnecessary biopsies.

A Deep Learning Method for Limited-View Intravascular Photoacoustic Image Reconstruction

2020 ◽  
Vol 10 (11) ◽  
pp. 2707-2713
Author(s):  
Zheng Sun ◽  
Xiangyang Yan
Keyword(s):  
Deep Learning ◽  
Image Reconstruction ◽  
Imaging Modality ◽  
Image Data ◽  
Structural Similarity ◽  
Simulated Image ◽  
Cross Sectional ◽  
Data Set ◽  
Reconstruction Methods ◽  
Under Sampling

Intravascular photoacoustic tomography (IVPAT) is a newly developed imaging modality in the interventional diagnosis and treatment of coronary artery diseases. Incomplete acoustic measurement caused by limitedview scanning of the detector in the vascular lumen results in under-sampling artifacts and distortion in the images reconstructed by using the standard reconstruction methods. A method for limited-view IVPAT image reconstruction based on deep learning is presented in this paper. A convolutional neural network (CNN) is constructed and trained with computer-simulated image data set. Then, the trained CNN is used to optimize the cross-sectional images of the vessel which are recovered from the incomplete photoacoustic measurements by using the standard time-reversal (TR) algorithm to obtain the images with the improved quality. Results of numerical demonstration indicate that the method can effectively reduce the image distortion and artifacts caused by the limited-view detection. Furthermore, it is superior to the compressed sensing (CS) method in recovering the unmeasured information of the imaging target with the structural similarity around 10% higher than CS reconstruction.

Sign in / Sign up

Export Citation Format

Share Document