Non-Invasive Diagnosis Model for Pancreatic Cystic Tumors Based on Machine Learning-Radiomics Using Contrast-Enhanced CT

2018 ◽  
Author(s):  
Xiaoyong Shen ◽  
Fan Yang ◽  
Pengfei Yang ◽  
Lei Xu ◽  
Modan Yang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Non Invasive ◽  
Contrast Enhanced Ct ◽  
Contrast Enhanced ◽  
Cystic Tumors ◽  
Enhanced Ct ◽  
Pancreatic Cystic Tumors ◽  
Diagnosis Model

scholarly journals CT diagnosis of ilioinguinal lymph node metastases in melanoma using radiological characteristics beyond size and asymmetry

BJS Open ◽  
2021 ◽  
Vol 5 (1) ◽  
Author(s):  
M J Wilkinson ◽  
H Snow ◽  
K Downey ◽  
K Thomas ◽  
A Riddell ◽  
...  
Keyword(s):  
Lymph Node ◽  
Diagnostic Accuracy ◽  
Multivariable Analysis ◽  
Ct Scans ◽  
Non Invasive ◽  
Poor Predictor ◽  
Contrast Enhanced Ct ◽  
Contrast Enhanced ◽  
Enhanced Ct ◽  
Node Metastases

Abstract Background Diagnosis of lymph node (LN) metastasis in melanoma with non-invasive methods is challenging. The aim of this study was to evaluate the diagnostic accuracy of six LN characteristics on CT in detecting melanoma-positive ilioinguinal LN metastases, and to determine whether inguinal LN characteristics can predict pelvic LN involvement. Methods This was a single-centre retrospective study of patients with melanoma LN metastases at a tertiary cancer centre between 2008 and 2016. Patients who had preoperative contrast-enhanced CT assessment and ilioinguinal LN dissection were included. CT scans containing significant artefacts obscuring the pelvis were excluded. CT scans were reanalysed for six LN characteristics (extracapsular spread (ECS), minimum axis (MA), absence of fatty hilum (FH), asymmetrical cortical nodule (CAN), abnormal contrast enhancement (ACE) and rounded morphology (RM)) and compared with postoperative histopathological findings. Results A total of 90 patients were included. Median age was 58 (range 23–85) years. Eighty-eight patients (98 per cent) had pathology-positive inguinal disease and, of these, 45 (51 per cent) had concurrent pelvic disease. The most common CT characteristics found in pathology-positive inguinal LNs were MA greater than 10 mm (97 per cent), ACE (80 per cent), ECS (38 per cent) and absence of RM (38 per cent). In multivariable analysis, inguinal LN characteristics on CT indicative of pelvic disease were RM (odds ratio (OR) 3.3, 95 per cent c.i. 1.2 to 8.7) and ECS (OR 4.2, 1.6 to 11.3). Cloquet’s node is known to be a poor predictor of pelvic spread. Pelvic LN disease was present in 50 per cent patients, but only 7 per cent had a pathology-positive Cloquet’s node. Conclusion Additional CT radiological characteristics, especially ECS and RM, may improve diagnostic accuracy and aid clinical decisions regarding the need for inguinal or ilioinguinal dissection.

scholarly journals Machine Learning for Detecting Early Infarction in Acute Stroke with Non–Contrast-enhanced CT

Radiology ◽  
2020 ◽  
Vol 294 (3) ◽  
pp. 638-644 ◽  
Author(s):  
Wu Qiu ◽  
Hulin Kuang ◽  
Ericka Teleg ◽  
Johanna M. Ospel ◽  
Sung Il Sohn ◽  
...  
Keyword(s):  
Machine Learning ◽  
Acute Stroke ◽  
Contrast Enhanced Ct ◽  
Contrast Enhanced ◽  
Enhanced Ct ◽  
Early Infarction

scholarly journals TO ASSESS THE ROLE OF DWI IN CHARACTERIZING THE RENAL NEOPLAMS INTO THE VARIOUS SUBTYPES USING THE ADC VALUES OF THE LESIONS AND FINDING ITS ACCURACY BY HISTOPATHOLOGICAL CORRELATION.

2020 ◽  
Vol 4 (10) ◽  
Author(s):  
Nitin Kumar Tripathi ◽  
Sanjay Kumar Tekam
Keyword(s):  
Renal Lesion ◽  
Medical College ◽  
Non Invasive ◽  
Histopathological Correlation ◽  
Contrast Enhanced Ct ◽  
Contrast Enhanced ◽  
Enhanced Ct ◽  
Adc Values

This prospective study was done in the Department of Radiodiagnosis at M.G.M. Medical College, & M.Y. Hospital, Indore. A total of 40 patients who were referred to our department with strong clinical suspicion of renal lesion and those diagnosed by ultrasonography or contrast enhanced CT scan having a renal mass on either of them, were subjected to a non-contrast. ADC values are expressed in 10-3 mm2/s. DW imaging with ADC values allows differentiation of clear cell, papillary and chromophobe subtypes of RCCs, suggesting that DW imaging may be a useful tool in the preoperative characterization of RCC. Hence, MRI with DWI in particular is a very valuable non invasive tool for the Identification, characterization and differentiation of renal lesions. Study Designed: Observational Study. Keywords: DWI, Renal, Neoplams, ADC, Histopathological & Correlation.

The potential of machine learning to predict postoperative pancreatic fistula based on preoperative, non-contrast-enhanced CT: A proof-of-principle study

Surgery ◽  
2020 ◽  
Vol 167 (2) ◽  
pp. 448-454 ◽  
Author(s):  
Patryk Kambakamba ◽  
Manoj Mannil ◽  
Paola E. Herrera ◽  
Philip C. Müller ◽  
Christoph Kuemmerli ◽  
...  
Keyword(s):  
Machine Learning ◽  
Pancreatic Fistula ◽  
Postoperative Pancreatic Fistula ◽  
Contrast Enhanced Ct ◽  
Contrast Enhanced ◽  
Enhanced Ct ◽  
Proof Of Principle

scholarly journals Application of CT-Based Radiomics in Discriminating Pancreatic Cystadenomas From Pancreatic Neuroendocrine Tumors Using Machine Learning Methods

2021 ◽  
Vol 11 ◽  
Author(s):  
Xuejiao Han ◽  
Jing Yang ◽  
Jingwen Luo ◽  
Pengan Chen ◽  
Zilong Zhang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Predictive Models ◽  
Pancreatic Neuroendocrine Tumors ◽  
Selection Methods ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Contrast Enhanced Ct ◽  
Contrast Enhanced ◽  
Enhanced Ct ◽  
Testing Group

ObjectivesThe purpose of this study aimed at investigating the reliability of radiomics features extracted from contrast-enhanced CT in differentiating pancreatic cystadenomas from pancreatic neuroendocrine tumors (PNETs) using machine-learning methods.MethodsIn this study, a total number of 120 patients, including 66 pancreatic cystadenomas patients and 54 PNETs patients were enrolled. Forty-eight radiomic features were extracted from contrast-enhanced CT images using LIFEx software. Five feature selection methods were adopted to determine the appropriate features for classifiers. Then, nine machine learning classifiers were employed to build predictive models. The performance of the forty-five models was evaluated with area under the curve (AUC), accuracy, sensitivity, specificity, and F1 score in the testing group.ResultsThe predictive models exhibited reliable ability of differentiating pancreatic cystadenomas from PNETs when combined with suitable selection methods. A combination of DC as the selection method and RF as the classifier, as well as Xgboost+RF, demonstrated the best discriminative ability, with the highest AUC of 0.997 in the testing group.ConclusionsRadiomics-based machine learning methods might be a noninvasive tool to assist in differentiating pancreatic cystadenomas and PNETs.

scholarly journals Computed Tomography-Based Radiomics in Predicting T Stage and Length of Esophageal Squamous Cell Carcinoma

2021 ◽  
Vol 11 ◽  
Author(s):  
Mingwei Yang ◽  
Panpan Hu ◽  
Minglun Li ◽  
Rui Ding ◽  
Yichun Wang ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Cluster Analysis ◽  
Tumor Length ◽  
Non Invasive ◽  
T Stage ◽  
Contrast Enhanced Ct ◽  
Contrast Enhanced ◽  
Enhanced Ct ◽  
Chi Squared ◽  
Tumor Extent

BackgroundBecause of the superficial and infiltrative spreading patterns of esophageal squamous cell carcinoma (ESCC), an accurate assessment of tumor extent is challenging using imaging-based clinical staging. Radiomics features extracted from pretreatment computed tomography (CT) or magnetic resonance imaging have shown promise in identifying tumor characteristics. Accurate staging is essential for planning cancer treatment, especially for deciding whether to offer surgery or radiotherapy (chemotherapy) in patients with locally advanced ESCC. Thus, this study aimed to evaluate the predictive potential of contrast-enhanced CT-based radiomics as a non-invasive approach for estimating pathological tumor extent in ESCC patients.MethodsPatients who underwent esophagectomy between October 2011 and September 2017 were retrospectively studied and included 116 patients with pathologically confirmed ESCC. Contrast-enhanced CT from the neck to the abdomen was performed in all patients during the 2 weeks before the operation. Radiomics features were extracted from segmentations, which were contoured by radiologists. Cluster analysis was performed to obtain clusters with similar radiomics characteristics, and chi-squared tests were used to assess differences in clinicopathological features and survival among clusters. Furthermore, a least absolute shrinkage and selection operator was performed to select radiomics features and construct a radiomics model. Receiver operating characteristic analysis was used to evaluate the predictive ability of the radiomics signatures.ResultsAll 116 ESCC patients were divided into two groups according to the cluster analysis. The chi-squared test showed that cluster-based radiomics features were significantly correlated with T stage (p = 0.0254) and tumor length (p = 0.0002). Furthermore, CT radiomics signatures exhibited favorable predictive performance for T stage (area under the curve [AUC] = 0.86, sensitivity = 0.77, and specificity = 0.87) and tumor length (AUC = 0.95, sensitivity = 0.92, and specificity = 0.91).ConclusionsCT contrast radiomics is a simple and non-invasive method that shows promise for predicting pathological T stage and tumor length preoperatively in ESCC patients and may aid in the accurate assessments of patients in combination with the existing examinations.

Sign in / Sign up

Export Citation Format

Share Document