scholarly journals Automatic detection of pituitary microadenoma from magnetic resonance imaging using deep learning algorithms

2020 ◽  
Author(s):  
Qingling Li ◽  
Yanhua Zhu ◽  
Minglin Chen ◽  
Ruomi Guo ◽  
Hao Liu ◽  
...  
Keyword(s):  
Magnetic Resonance ◽  
Diagnostic Accuracy ◽  
Mr Images ◽  
Training Set ◽  
Cad System ◽  
Pituitary Microadenoma ◽  
Diagnosis Accuracy ◽  
Validation Set ◽  
Primary Health Care Institutions ◽  
Aided Diagnosis

Abstract The risks of misdiagnosed pituitary microadenoma is high. We designed a convolutional neural network (CNN) based computer-aided diagnosis (CAD) system to retrospectively diagnose patients with pituitary microadenoma. A total 5,540 pituitary magnetic resonance (MR) images from 1,108 participants were recruited. MRI images were randomly stratified into non-overlapping cohorts (training set, validation set and test set) to establish five different CNN models. The best CNN model is the ResNet with a diagnostic accuracy of 94%, which outperforms the diagnosis accuracy of our radiologists (64%-85%). The accuracy of our CAD system is further confirmed in additional MR datasets. The diagnostic accuracy of our ResNet model is comparable to the proficiency of a radiologist with 5-10 years’ experience. In summary, this is the first report showing that the CAD system is a viable tool for diagnosing pituitary microadenoma. CAD system is applicable to radiology departments, especially in primary health care institutions.

scholarly journals Automatic detection of pituitary microadenoma from magnetic resonance imaging using deep learning algorithms

2021 ◽  
Author(s):  
Qingling Li ◽  
Yanhua Zhu ◽  
Minglin Chen ◽  
Ruomi Guo ◽  
Qingyong Hu ◽  
...  
Keyword(s):  
Deep Learning ◽  
Diagnostic Accuracy ◽  
Network Models ◽  
Clinical Workflow ◽  
Neural Network Models ◽  
Test Dataset ◽  
Cad System ◽  
Pituitary Microadenoma ◽  
Primary Health ◽  
Primary Health Care Institutions

Pituitary microadenoma (PM) is often difficult to detect by MR imaging alone. We employed a computer-aided PM diagnosis (PM-CAD) system based on deep learning to assist radiologists in clinical workflow. We enrolled 1,228 participants and stratified into 3 non-overlapping cohorts for training, validation and testing purposes. Our PM-CAD system outperformed 6 existing established convolutional neural network models for detection of PM. In test dataset, diagnostic accuracy of PM-CAD system was comparable to radiologists with > 10 years of professional expertise (94% versus 95%). The diagnostic accuracy in internal and external dataset was 94% and 90%, respectively. Importantly, PM-CAD system detected the presence of PM that had been previously misdiagnosed by radiologists. This is the first report showing that PM-CAD system is a viable tool for detecting PM. Our results suggest that PM-CAD system is applicable to radiology departments, especially in primary health care institutions.

scholarly journals Development and Validation of a Deep Learning Algorithm to Automatic Detection of Pituitary Microadenoma From MRI

2021 ◽  
Vol 8 ◽  
Author(s):  
Qingling Li ◽  
Yanhua Zhu ◽  
Minglin Chen ◽  
Ruomi Guo ◽  
Qingyong Hu ◽  
...  
Keyword(s):  
Deep Learning ◽  
Diagnostic Accuracy ◽  
Roc Curve ◽  
Predictive Value ◽  
Computer Aided Diagnosis ◽  
Validation Dataset ◽  
Cad System ◽  
Pituitary Microadenoma ◽  
Computer Aided ◽  
Aided Diagnosis

Background: It is often difficult to diagnose pituitary microadenoma (PM) by MRI alone, due to its relatively small size, variable anatomical structure, complex clinical symptoms, and signs among individuals. We develop and validate a deep learning -based system to diagnose PM from MRI.Methods: A total of 11,935 infertility participants were initially recruited for this project. After applying the exclusion criteria, 1,520 participants (556 PM patients and 964 controls subjects) were included for further stratified into 3 non-overlapping cohorts. The data used for the training set were derived from a retrospective study, and in the validation dataset, prospective temporal and geographical validation set were adopted. A total of 780 participants were used for training, 195 participants for testing, and 545 participants were used to validate the diagnosis performance. The PM-computer-aided diagnosis (PM-CAD) system consists of two parts: pituitary region detection and PM diagnosis. The diagnosis performance of the PM-CAD system was measured using the receiver operating characteristics (ROC) curve and area under the ROC curve (AUC), calibration curve, accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and F1-score.Results: Pituitary microadenoma-computer-aided diagnosis system showed 94.36% diagnostic accuracy and 98.13% AUC score in the testing dataset. We confirm the robustness and generalization of our PM-CAD system, the diagnostic accuracy in the internal dataset was 96.50% and in the external dataset was 92.26 and 92.36%, the AUC was 95.5, 94.7, and 93.7%, respectively. In human-computer competition, the diagnosis performance of our PM-CAD system was comparable to radiologists with >10 years of professional expertise (diagnosis accuracy of 94.0% vs. 95.0%, AUC of 95.6% vs. 95.0%). For the misdiagnosis cases from radiologists, our system showed a 100% accurate diagnosis. A browser-based software was designed to assist the PM diagnosis.Conclusions: This is the first report showing that the PM-CAD system is a viable tool for detecting PM. Our results suggest that the PM-CAD system is applicable to radiology departments, especially in primary health care institutions.

scholarly journals Clinical predictive efficacy of C-reactive protein for diagnosing infectious complications after gastric surgery

2020 ◽  
Vol 13 ◽  
pp. 175628482093654
Author(s):  
Jinyao Shi ◽  
Zhouqiao Wu ◽  
Qi Wang ◽  
Yan Zhang ◽  
Fei Shan ◽  
...  
Keyword(s):  
Diagnostic Accuracy ◽  
Anastomotic Leakage ◽  
Enhanced Recovery ◽  
Gastric Surgery ◽  
Infectious Complications ◽  
C Reactive Protein ◽  
Characteristic Analysis ◽  
Training Set ◽  
Reactive Protein ◽  
Validation Set

Background: With the popularization of Enhanced Recovery After Surgery (ERAS), identifying patients with complications before discharging becomes important. This study aimed to explore the efficacy of C-reactive protein (CRP) in predicting infectious complications after gastrectomy. Methods: Patients with gastric cancer who underwent gastrectomy at Beijing Cancer Hospital from March 2017 to April 2018 were enrolled in the training set. Complications were prospectively registered. Receiver operating characteristic analysis was performed to assess the diagnostic accuracy of CRP via evaluating the area under the curve (AUC). Patients who had CRP tested on postoperative day (POD) 5 and accepted gastrectomy from April to December 2018 were included in the validation set to validate the cut-off value of CRP obtained from the training set. Results: A total of 350 patients were included (263 patients in the training set and 87 patients in the validation set). Out of these, 24 patients were diagnosed with infectious complications and 17 patients had anastomotic leakage in the training set. The CRP level on POD5 had superior diagnostic accuracy for infectious complications with an AUC of 0.81. The cut-off value of CRP on POD5 at 166.65 mg/L yielded 93% specificity and 97.2% negative predict value (NPV); For anastomotic leakage, the AUC of CRP on POD5 was 0.81. Using the cut-off value of CRP at 166.65 mg/L on POD5 achieved 92% specificity and 98.6% NPV. The optimal cut-off value (CRP 166.65 mg/L on POD5) was validated in the validation set. It achieved 97.5% specificity and 94.0% NPV for infectious complications, and 97.6% specificity and 96.4% NPV for anastomotic leakage. Conclusion: CRP is a reliable predictive marker for the diagnosis of inflammatory complications following gastric surgery. However, this study was based on preliminary data. The validity of this data needs confirmation by a larger number of cases.

scholarly journals The role of red cell distribution width in the differential diagnosis of iron deficiency anemia and non-transfusion dependent thalassemia patients

2018 ◽  
Vol 10 (3) ◽  
Author(s):  
Pokpong Piriyakhuntorn ◽  
Adisak Tantiworawit ◽  
Thanawat Rattanathammethee ◽  
Chatree Chai-Adisaksopha ◽  
Ekarat Rattarittamrong ◽  
...  
Keyword(s):  
Diagnostic Accuracy ◽  
Predictive Value ◽  
Area Under The Curve ◽  
Microcytic Anemia ◽  
Deficiency Anemia ◽  
Operating Characteristics ◽  
Training Set ◽  
Distribution Width ◽  
Validation Set ◽  
Sensitivity Specificity

This study aims to find the cut-off value and diagnostic accuracy of the use of RDW as initial investigation in enabling the differentiation between IDA and NTDT patients. Patients with microcytic anemia were enrolled in the training set and used to plot a receiving operating characteristics (ROC) curve to obtain the cut-off value of RDW. A second set of patients were included in the validation set and used to analyze the diagnostic accuracy. We recruited 94 IDA and 64 NTDT patients into the training set. The area under the curve of the ROC in the training set was 0.803. The best cut-off value of RDW in the diagnosis of NTDT was 21.0% with a sensitivity and specificity of 81.3% and 55.3% respectively. In the validation set, there were 34 IDA and 58 NTDT patients using the cut-off value of >21.0% to validate. The sensitivity, specificity, positive predictive value and negative predictive value were 84.5%, 70.6%, 83.1% and 72.7% respectively. We can therefore conclude that RDW >21.0% is useful in differentiating between IDA and NTDT patients with high diagnostic accuracy

scholarly journals Development and Validation of a Scoring System for Early Diagnosis of Malignant Pleural Effusion Based on a Nomogram

2021 ◽  
Vol 11 ◽  
Author(s):  
Aihua Wu ◽  
Zhigang Liang ◽  
Songbo Yuan ◽  
Shanshan Wang ◽  
Weidong Peng ◽  
...  
Keyword(s):  
Pleural Effusion ◽  
Diagnostic Accuracy ◽  
Prediction Model ◽  
Scoring System ◽  
Malignant Pleural Effusion ◽  
Diagnostic Performance ◽  
External Validation ◽  
Training Set ◽  
Internal Validation ◽  
Validation Set

BackgroundThe diagnostic value of clinical and laboratory features to differentiate between malignant pleural effusion (MPE) and benign pleural effusion (BPE) has not yet been established.ObjectivesThe present study aimed to develop and validate the diagnostic accuracy of a scoring system based on a nomogram to distinguish MPE from BPE.MethodsA total of 1,239 eligible patients with PE were recruited in this study and randomly divided into a training set and an internal validation set at a ratio of 7:3. Logistic regression analysis was performed in the training set, and a nomogram was developed using selected predictors. The diagnostic accuracy of an innovative scoring system based on the nomogram was established and validated in the training, internal validation, and external validation sets (n = 217). The discriminatory power and the calibration and clinical values of the prediction model were evaluated.ResultsSeven variables [effusion carcinoembryonic antigen (CEA), effusion adenosine deaminase (ADA), erythrocyte sedimentation rate (ESR), PE/serum CEA ratio (CEA ratio), effusion carbohydrate antigen 19-9 (CA19-9), effusion cytokeratin 19 fragment (CYFRA 21-1), and serum lactate dehydrogenase (LDH)/effusion ADA ratio (cancer ratio, CR)] were validated and used to develop a nomogram. The prediction model showed both good discrimination and calibration capabilities for all sets. A scoring system was established based on the nomogram scores to distinguish MPE from BPE. The scoring system showed favorable diagnostic performance in the training set [area under the curve (AUC) = 0.955, 95% confidence interval (CI) = 0.942–0.968], the internal validation set (AUC = 0.952, 95% CI = 0.932–0.973), and the external validation set (AUC = 0.973, 95% CI = 0.956–0.990). In addition, the scoring system achieved satisfactory discriminative abilities at separating lung cancer-associated MPE from tuberculous pleurisy effusion (TPE) in the combined training and validation sets.ConclusionsThe present study developed and validated a scoring system based on seven parameters. The scoring system exhibited a reliable diagnostic performance in distinguishing MPE from BPE and might guide clinical decision-making.

Detection of Artery Regions in Lower Extremity Arteries from Non-Enhanced MR Imaging Based on Particle Filter Algorithms

2013 ◽  
Vol 17 (2) ◽  
pp. 318-323 ◽  
Author(s):  
Yuiko Koga ◽  
◽  
Akiyoshi Yamamoto ◽  
Hyoungseop Kim ◽  
Joo Kooi Tan ◽  
...  
Keyword(s):  
Particle Filter ◽  
Operating Characteristic ◽  
Arterial Disease ◽  
Mr Images ◽  
Standard Data ◽  
Cad System ◽  
Peripheral Arterial ◽  
Blood Imaging ◽  
Filter Algorithms ◽  
Aided Diagnosis

Recently, the arteries sclerosis obliterans (ASO) or called peripheral arterial disease (PAD) typically caused by chronic ischemia of limbs increases remarkably. As one of the diagnosis methods, the image diagnosis methods such as MR image are applied in medical fields. In this paper, we propose a vascular extraction method using fresh blood imaging (FBI) method, as well as apply it to computer aided diagnosis (CAD) system. Especially, to prevent the spread outside of the region and improve the segment accuracy of peripheral artery areas, we introduce particle filter algorithms. We performed our method on automatic artery regions detection using non-enhanced MR images. Furthermore, we compared the extracted results to gold standard data and analyzed accuracy by receiver operating characteristic (ROC). The effectiveness of our proposed method and satisfactory of its detected accuracy were confirmed.

Accuracy of endorectal Magnetic Resonance Imaging (MRI) and Dynamic Contrast Enhanced-MRI (DCE-MRI) in the preoperative local staging of prostate cancer

2012 ◽  
Vol 79 (2) ◽  
pp. 116-122 ◽  
Author(s):  
Alessandro Baccos ◽  
Riccardo Schiavina ◽  
Ziv Zukerman ◽  
Fiorenza Busato ◽  
Caterina Gaudianol ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Prostate Cancer ◽  
Magnetic Resonance ◽  
Radical Prostatectomy ◽  
Diagnostic Accuracy ◽  
Resonance Imaging ◽  
Mr Images ◽  
Dce Mri ◽  
Local Staging ◽  
Magnetic Resonance Imaging Mri

Background The proper management of newly diagnosed prostate cancer (PCa) requires the choice of the appropriate treatment plan. A crucial factor is the accurate evaluation of the tumor local extension. The Magnetic Resonance Imaging (MRI) plays an important role in the local staging of prostate cancer, although its use in clinical practice is widely debated. Therefore, the purpose of our study was to evaluate the diagnostic accuracy of T2-weighted MR imaging in association with DCE-MRI, performed using an endorectal coil, in preoperative local staging of patients with prostate cancer, by using the histopathologic findings as the reference standard. Materials and Methods From April 2010 to May 2011, 65 patients (mean age, 65 years; range, 51–77 years) with clinical localized PCa, underwent radical prostatectomy at our institution, performed by 2 experienced surgeons. All patients were prospectively evaluated with eMRI in association with DCE-MRI prior to radical prostatectomy. In all patients MRI was performed at least 6 weeks after biopsy and within 2 weeks before Radical Prostatectomy (RP). Histologic analysis was our diagnostic “gold standard”. To ensure that the histopathological findings matched with MR images, the assessment of radiological images and the RP specimens were performed dividing the prostate in 14 regions. Results First, we performed a “per-patient” analysis, considering the entire prostate as a single region. Then, we performed a “per-emigland” analysis, finally a “per-region” analysis. The sensitivity, specificity, PPV, NPV and AUC in predicting ECE in the analysis “per-emigland” were respectively 66.7, 95.7, 66.7, 95.7, 0.824. The evaluation of SVI reported similar results: 62.5, 97.5, 62.5, 97.5, 0.797. DCE-MRI did not improve the diagnostic accuracy of T1-T2-weighted MR images in the evaluation of ECE or SVI. Conclusions T1-, T2-weighted MRI adds important information regarding the preoperative local staging of PCa. DCE-MRI does not improve the diagnostic accuracy of MRI in the local staging of PCa.

scholarly journals Utility of diffusion-weighted and contrast-enhanced magnetic resonance imaging in diagnosing and differentiating between high- and low-grade uterine endometrial stromal sarcoma

2019 ◽  
Vol 19 (1) ◽  
Author(s):  
Yen-Ling Huang ◽  
Shir-Hwa Ueng ◽  
Kueian Chen ◽  
Yu-Ting Huang ◽  
Hsin-Ying Lu ◽  
...  
Keyword(s):  
Magnetic Resonance ◽  
Diagnostic Accuracy ◽  
Endometrial Stromal Sarcoma ◽  
Low Grade ◽  
Imaging Features ◽  
High Grade ◽  
Mr Images ◽  
Stromal Sarcoma ◽  
Diffusion Weighted ◽  
Contrast Enhanced

Abstract Background Endometrial stromal sarcoma (ESS) is a rare uterine malignancy that features different prognoses for its high- and low-grade subtypes. We investigated the diagnostic accuracy of magnetic resonance (MR) imaging in diagnosing and differentiating between high- and low-grade ESS. Methods We retrospectively reviewed the preoperative pelvic MR images of consecutive patients who received histologically confirmed diagnoses of high-grade ESS (n = 11) and low-grade ESS (n = 9) and T2-hyperintense leiomyoma (n = 16). Two radiologists independently evaluated imaging features in T1-, T2-, and diffusion-weighted and contrast-enhanced MR images. Statistical analysis included Mann-Whitney tests and Fisher’s exact test, with sensitivity, specificity and diagnostic accuracy of imaging features. Results High-grade ESS was associated with significantly more extensive necrosis and hemorrhage and distinct feather-like enhancement compared with low-grade ESS (P < .05 for all). The feather-like enhancement pattern yielded a diagnostic accuracy of 95%, sensitivity of 91%, and specificity of 100% in differentiating high-grade from low-grade ESS. This imaging characteristic was significantly superior to the necrosis (80%, P = .033) or hemorrhage (75%, P = .007). Both high- and low-grade ESS demonstrated T2 hypointense bands, marginal nodules, intratumoral nodules, and worm-like intra-myometrial nodules, and their tumor apparent diffusion coefficient (ADC) values were significantly lower than those of T2-hyperintense leiomyomas (P < .001). Conclusions Diffusion-weighted MR imaging is useful in diagnosing ESS against T2-hyperintense leiomyomas, and contrast enhancement aids in further differentiating between high- and low-grade ESS.

scholarly journals Knee Osteoarthritis Classification Using 3D CNN and MRI

2021 ◽  
Vol 11 (11) ◽  
pp. 5196
Author(s):  
Carmine Guida ◽  
Ming Zhang ◽  
Juan Shan
Keyword(s):  
Mr Images ◽  
X Ray ◽  
3 Dimensional ◽  
Knee Oa ◽  
Grade Classification ◽  
Kellgren And Lawrence ◽  
Diagnosis Accuracy ◽  
Validation Set ◽  
3D Cnn ◽  
Ray Methods

Osteoarthritis (OA) is the most common form of arthritis and can often occur in the knee. While convolutional neural networks (CNNs) have been widely used to study medical images, the application of a 3-dimensional (3D) CNN in knee OA diagnosis is limited. This study utilizes a 3D CNN model to analyze sequences of knee magnetic resonance (MR) images to perform knee OA classification. An advantage of using 3D CNNs is the ability to analyze the whole sequence of 3D MR images as a single unit as opposed to a traditional 2D CNN, which examines one image at a time. Therefore, 3D features could be extracted from adjacent slices, which may not be detectable from a single 2D image. The input data for each knee were a sequence of double-echo steady-state (DESS) MR images, and each knee was labeled by the Kellgren and Lawrence (KL) grade of severity at levels 0–4. In addition to the 5-category KL grade classification, we further examined a 2-category classification that distinguishes non-OA (KL ≤ 1) from OA (KL ≥ 2) knees. Clinically, diagnosing a patient with knee OA is the ultimate goal of assigning a KL grade. On a dataset with 1100 knees, the 3D CNN model that classifies knees with and without OA achieved an accuracy of 86.5% on the validation set and 83.0% on the testing set. We further conducted a comparative study between MRI and X-ray. Compared with a CNN model using X-ray images trained from the same group of patients, the proposed 3D model with MR images achieved higher accuracy in both the 5-category classification (54.0% vs. 50.0%) and the 2-category classification (83.0% vs. 77.0%). The result indicates that MRI, with the application of a 3D CNN model, has greater potential to improve diagnosis accuracy for knee OA clinically than the currently used X-ray methods.

scholarly journals Optimized Resolution-Oriented Many-to-One Intensity Standardization Method for Magnetic Resonance Images

2019 ◽  
Vol 9 (24) ◽  
pp. 5531
Author(s):  
Yuan Gao ◽  
Yuanyuan Wang ◽  
Jinhua Yu
Keyword(s):  
Magnetic Resonance ◽  
Visual Information ◽  
Signal To Noise Ratio ◽  
Structural Similarity ◽  
Computer Aided Diagnosis ◽  
Magnetic Resonance Images ◽  
Image Resolution ◽  
Mr Images ◽  
Computer Aided ◽  
Aided Diagnosis

With the development of big data, Radiomics and deep-learning methods based on magnetic resonance (MR) images, it is necessary to conduct large databases containing MR images from multiple centers. Having huge intensity distribution differences among images reduced or even eliminated, robust computer-aided diagnosis models could be established. Therefore, an optimized intensity standardization model is proposed. The network structure, loss function, and data input strategy were optimized to better avoid the image resolution loss during transformation. The experimental dataset was obtained from five MR scanners located in four hospitals and was divided into nine groups based on the imaging parameters, during which 9152 MR images from 499 participants were collected. Experiments show the superiority of the proposed method to the previously proposed unified model in resolution metrics including the peak signal-to-noise ratio, structural similarity, visual information fidelity, universal quality index, and image fidelity criterion. Another experiment further shows the advantage of the proposed method in increasing the effectiveness of following computer-aided diagnosis models by better preservation of MR image details. Moreover, the advantage over conventional standardization methods are also shown. Thus, MR images from different centers can be standardized using the proposed method, which will facilitate numerous data-driven medical imaging studies.

Sign in / Sign up

Export Citation Format

Share Document