CT Imaging Features Research Articles

A deep fuzzy model for diagnosis of COVID-19 from CT images

From early 2020, a novel coronavirus disease pneumonia has shown a global “pandemic” trend at an extremely fast speed. Due to the magnitude of its harm, it has become a major global public health event. In the face of dramatic increase in the number of patients with COVID-19, the need for quick diagnosis of suspected cases has become particularly critical. Therefore, this paper constructs a fuzzy classifier, which aims to detect infected subjects by observing and analyzing the CT images of suspected patients. Firstly, a deep learning algorithm is used to extract the low-level features of CT images in the COVID-CT dataset. Subsequently, we analyze the extracted feature information with attribute reduction algorithm to obtain features with high recognition. Then, some key features are selected as the input for the fuzzy diagnosis model to the training model. Finally, several images in the dataset are used as the test set to test the trained fuzzy classifier. The obtained accuracy rate is 94.2%, and the F1-score is 93.8%. Experimental results show that, compared with the deep learning diagnosis methods widely used in medical image analysis, the proposed fuzzy model improves the accuracy and efficiency of diagnosis, which consequently helps to curb the spread of COVID-19.

Analysis of CT and MRI Imaging Characteristics of Spinal Invasive Hemangioma

Analysis of CT and MRI Imaging Characteristics of Spinal Invasive Hemangioma

Multiphase Dual-Energy Spectral CT-Based Deep Learning Method for the Noninvasive Prediction of Head and Neck Lymph Nodes Metastasis in Patients With Papillary Thyroid Cancer.

PurposeTo develop deep learning (DL) models based on multiphase dual-energy spectral CT for predicting lymph nodes metastasis preoperatively and noninvasively in papillary thyroid cancer patients.MethodsA total of 293 lymph nodes from 78 papillary thyroid cancer patients who underwent dual-energy spectral CT before lymphadenectomy were enrolled in this retrospective study. The lymph nodes were randomly divided into a development set and an independent testing set following a 4:1 ratio. Four single-modality DL models based on CT-A model, CT-V model, Iodine-A model and Iodine-V model and a multichannel DL model incorporating all modalities (Combined model) were proposed for the prediction of lymph nodes metastasis. A CT-feature model was also built on the selected CT image features. The model performance was evaluated with respect to discrimination, calibration and clinical usefulness. In addition, the diagnostic performance of the Combined model was also compared with four radiologists in the independent test set.ResultsThe AUCs of the CT-A, CT-V, Iodine-A, Iodine-V and CT-feature models were 0.865, 0.849, 0.791, 0.785 and 0.746 in the development set and 0.830, 0.822, 0.744, 0.739 and 0.732 in the testing set. The Combined model had outperformed the other models and achieved the best performance with AUCs yielding 0.890 in the development set and 0.865 in the independent testing set. The Combined model showed good calibration, and the decision curve analysis demonstrated that the net benefit of the Combined model was higher than that of the other models across the majority of threshold probabilities. The Combined model also showed noninferior diagnostic capability compared with the senior radiologists and significantly outperformed the junior radiologists, and the interobserver agreement of junior radiologists was also improved after artificial intelligence assistance.ConclusionThe Combined model integrating both CT images and iodine maps of the arterial and venous phases showed good performance in predicting lymph nodes metastasis in papillary thyroid cancer patients, which could facilitate clinical decision-making.

Clinical and molecular pathological features of bronchopulmonary large cell neuroendocrine carcinoma

Objective: To investigate the clinical manifestations, imaging, pathological and molecular features of bronchopulmonary large-cell neuroendocrine carcinoma (LCNEC). Methods: The clinical data of 216 LCNEC patients in the First Affiliated Hospital of Zhengzhou University from 2011 to 2021 were analyzed retrospectively. The clinical manifestations, tumor location and size, characteristics of CT images, immunohistochemical and molecular pathological features were analyzed and compared with 115 cases of mixed small cell carcinoma (M-SCLC) diagnosed in the same period. Results: Among the 216 LCNEC patients, there were 190 males and 26 females, with a median age of 65 years. The first symptoms of the patients were mainly cough (106 cases, 49.1%) and bloody sputum (48 cases, 22.2%). The median tumor length were 4.7cm, including 55 cases of nodular type (25.5%) and 161 cases of mass-forming type (74.5%). CT imaging results showed that LCNEC lesions had soft tissue density, and the proportion of slight enhancement lesions was significantly lower than that in M-SCLC group (52.3% vs 74.8%, P<0.001). In contrast, the proportion of necrosis (87.0% vs 58.3%, P<0.001) and calcification (26.9% vs 2.6%, P<0.001) in LCNEC patients was significantly higher than that in M-SCLC group. Immunohistochemical results showed that the positive rate of CK in LCNEC was significantly higher than that in M-SCLC (99.0 % vs 90.5%, P<0.05), while the positive rate of TTF-1 was significantly lower than that in M-SCLC (51.6% vs 67.0%, P<0.05). In LCNEC group, the proportion of patients with Ki-67 positive index between 50% and 80% was significantly higher than that of M-SCLC (41.2% vs 25.2%), while the proportion between 80% and 100% was lower than that of M-SCLC (51.9% vs 72.2%). There was no significant difference in the positive rates of CD56 (91.7% vs 94.6%, P=0.336), Syn (83.8% vs 84.7%, P=0.838) and CgA (54.8% vs 50.0%, P=0.632) in both tumor types. Molecular pathology results showed that frequent mutatios were TP53 (54.5%), RB1 (36.4%), KEAP1 (18.2%), MYC(18.2%), and PTEN (14.3%), and the rate of tumor mutation burden which is more than 25 mutation/Mb was 27.3%. Conclusions: LCNEC lacks specific clinical manifestations. CT imaging is powerful in distinguishing LCNEC from M-SCLC. LCNEC contains a specific mutation spectrum. Pathology combined with immunohistochemical staining is still the gold standard for LCNEC diagnosis, and the differentiation from M-SCLC mainly depends on cell size and nuclear chromatin pattern with light microscopy.

Preoperative prediction of intra-tumoral tertiary lymphoid structures based on CT in hepatocellular cancer

PurposeIntra-tumoral tertiary lymphoid structures (TLSs) are associated with a favorable prognosis for patients with hepatocellular carcinoma (HCC). We aimed to identify image features related to TLSs and develop a nomogram for preoperative noninvasive prediction of intra-tumoral TLSs. MethodsThis retrospective study enrolled patients with HCC who underwent contrast-enhanced computed tomography before surgery between January 2014 and September 2020. Two radiologists retrospectively and independently reviewed the CT imaging features, and interobserver agreement was assessed. Univariable and multivariable logistic regression analyses were applied to investigate clinical laboratory data and imaging features related to TLSs. A regression-based predictive model and nomogram were constructed using the identified predictors. Nomogram diagnostic performance was assessed with the area under the receiver operating characteristic curve (AUC) and calibration curves, and validated using 5-fold cross-validation. ResultsNinety-three of the 142 HCCs were TLS + HCCs. Multivariable analyses identified intratumor arteries (odds ratio [OR]: 0.23; 95% confidence interval [CI]: 0.07–0.63; p = 0.007), intratumor hemorrhage (OR: 0.08; 95% CI: 0.01–0.50; p = 0.012), positive HBsAg or HCVAB status (OR: 4.52; 95% CI: 1.65–13.29; p = 0.004), platelet count (≥186.5 × 109 /L, OR: 0.38; 95% CI: 0.16–0.86; p = 0.022), and aspartate transaminase level (≥33.2 IU/l, OR: 0.24; 95% CI: 0.09–0.59; p = 0.003) as independent predictors of intra-tumoral TLSs. AUC of the regression-based model was 0.79 (95% CI:0.72–0.86) and average AUC at 5-fold cross-validation was 0.75 (95% CI: 0.71–0.80). ConclusionsCT-based nomogram is promising for preoperative prediction of intra-tumoral TLS in HCC.

The feasibility analysis of multi-slice spiral CT features of gastric stromal tumors in predicting pathological National Institutes of Health risk classification

Objective: To investigate the feasibility of multi-slice spiral CT(MSCT) imaging features of gastric stromal tumor (GST) in predicting pathological NIH risk classification, providing imaging basis for patients with GST before treatment. Methods: The clinical and CT imaging data of 504 patients(506 GST lesions), 259males and 245 females, aged from 13 to 85(60±11) years, with GST confirmed by surgery and pathology collected in the Zhongshan Hospital Affiliated to Fudan University and the Affiliated TCM Hospital of Southwest Medical University. According to pathological NIH risk classification, 506 lesions were divided into low risk group (very low and low risk degree, 277 lesions) and high risk group (medium and high risk degree, 229 lesions).Clinical data and imaging characteristics were compared between two groups. Multivariate logistic regression analysis was performed to screen out independent risk factors for statistically significant imaging indicators. Receiver operating curve (ROC) was used to evaluate the predictive value of tumor length for risk classification. Resulst: Between low risk group and high risk group,there were significant differences in gender(male/female:131/146 vs 129/100), gastrointestinal bleeding(present/absent:39/238 vs 59/170), morphology(regular/Irregular:218/59 vs 95/134), calcification(present/absent:36/241 vs 53/176), degree of necrosis(0°/Ⅰ°/Ⅱ°/Ⅲ°:197/61/16/3 vs 58/98/32/41), ulceration(present/absent:32/245 vs 94/135), growth pattern(endophytic/exophytic/mixed:102/105/70 vs 44/98/87), tumor location(fundus/cardia/body/angle/antrum:98/7/135/12/25 vs 98/6/114/5/6), feeding artery(present/absent:32/245 vs 104/125), vascular enhancement(present/absent:19/258 vs 88/141), effusion of around the disease(present/absent:0/277 vs 13/216), positive sign of fat around the disease(present/absent:0/277 vs 30/199),maximum long diameter[2.82(2.04,3.80) cm vs 5.93(4.06,8.29) cm] and short diameter [2.31(1.60,2.88) cm vs 4.40(3.21,6.37) cm]of tumor(all P<0.05).The maximum long diameter of tumor (OR=2.08,95%CI:1.35-3.20) and ulceration positive(OR=2.01,95%CI:1.03-3.92) were independent risk factors of risk classification(all P<0.05).Gastric antrum was used as the reference for tumor location, gastric fundus(OR=7.77,95%CI:2.00-30.24) and gastric body(OR=3.93,95%CI:1.03-15.01) were independent risk factors of risk classification(all P<0.05).The area under curve(AUC) of the maximum long diameter of tumor for predicting risk classification was 0.87, and the optimal critical value, sensitivity and specificity were 4.98cm, 62.9% and 95.3% respectively. Conclusions: MSCT image features of GST had certain characteristics. MSCT has certain predictive value for pathological NIH risk classification of GST, which can provide certain imaging basis for patients before treatment.

Impact of improved spatial resolution on radiomic features using photon-counting-detector CT.

Radiomics is a promising mathematical tool for characterizing disease and predicting clinical outcomes from radiological images such as CT. Photon-counting-detector (PCD) CT provides improved spatial resolution and dose efficiency relative to conventional energy-integrating-detector CT systems. Since improved spatial resolution enables visualization of smaller structures and more details that are not typically visible at routine resolution, it has a direct impact on textural features in CT images. Therefore, it is of clinical interest to quantify the impact of the improved spatial resolution on calculated radiomic features and, consequently, on sample classification. In this work, organic samples (zucchini, onions, and oranges) were scanned on both clinical PCD-CT and EID-CT systems at two dose levels. High-resolution PCD-CT and routine-resolution EID-CT images were reconstructed using a dedicated sharp kernel and a routine kernel, respectively. The noise in each image was quantified. Fourteen radiomic features of relevance were calculated in each image for each sample and compared between the two scanners. Radiomic features were plotted pairwise to evaluate the resulting cluster separation of the samples by their type between PCD-CT and EID-CT. Thirteen out of 14 studied radiomic features were notably changed by the improved resolution of the PCD-CT system, and the cluster separation was better when assessing features derived from PCD-CT. These results show that features derived from high-resolution PCD-CT, which are subject to higher noise compared to EID-CT, may impact radiomics-based clinical decision making.

Research on Detection and Diagnosis Technology of Subarachnoid Hemorrhage Based on New Association Classification Algorithm

Subarachnoid hemorrhage (SAH) is one of the most common cerebrovascular emergencies, which can lead to serious consequences. Spontaneous subarachnoid hemorrhage accounts for about 15% of acute cerebrovascular accidents. Among them, spontaneous subarachnoid hemorrhage caused by rupture of an intracranial aneurysm or vascular malformation is more common, accounting for about 85%. Therefore, it is very important to detect the early symptoms of subarachnoid hemorrhage through reasonable means and to carry out appropriate clinical intervention and treatment. With the development of imaging technology, computed tomography angiography (CTA) is widely used in clinical practice. However, the accuracy of manual recognition of CT images is not very high, and the efficiency is low. The emergence of data mining technology is gradually solving this problem. In this paper, we introduce and summarize the development of data mining, domestic and foreign research progress, the application status of data mining in the medical field, and the main technologies and methods of data mining. We study the application of association rule extraction technology in data mining in the medical field. The Apriori algorithm for finding frequent item sets in association rule extraction and its series of improved algorithms are studied and finally combined the characteristics of medical CT images, an image mining method of association rules based on the gray-level cooccurrence matrix is proposed. Based on the FP-growth algorithm, the NCFP-growth algorithm based on association rules is proposed and compared with the mining effect of several other algorithms. The proposed algorithm achieves a classification accuracy of above 90%, which is higher than the Apriori algorithm and its improved variations.

A Clinical-Radiomic Model for Predicting Indocyanine Green Retention Rate at 15 Min in Patients With Hepatocellular Carcinoma.

Purpose:The indocyanine green retention rate at 15 min (ICG-R15) is of great importance in the accurate assessment of hepatic functional reserve for safe hepatic resection. To assist clinicians to evaluate hepatic functional reserve in medical institutions that lack expensive equipment, we aimed to explore a novel approach to predict ICG-R15 based on CT images and clinical data in patients with hepatocellular carcinoma (HCC).MethodsIn this retrospective study, 350 eligible patients were enrolled and randomly assigned to the training cohort (245 patients) and test cohort (105 patients). Radiomics features and clinical factors were analyzed to pick out the key variables, and based on which, we developed the random forest regression, extreme gradient boosting regression (XGBR), and artificial neural network models for predicting ICG-R15, respectively. Pearson's correlation coefficient (R) was adopted to evaluate the performance of the models.ResultsWe extracted 660 CT image features in total from each patient. Fourteen variables significantly associated with ICG-R15 were picked out for model development. Compared to the other two models, the XGBR achieved the best performance in predicting ICG-R15, with a mean difference of 1.59% (median, 1.53%) and an R-value of 0.90. Delong test result showed no significant difference in the area under the receiver operating characteristic (AUROCs) for predicting post hepatectomy liver failure between actual and estimated ICG-R15.ConclusionThe proposed approach that incorporates the optimal radiomics features and clinical factors can allow for individualized prediction of ICG-R15 value of patients with HCC, regardless of the specific equipment and detection reagent (NO. ChiCTR2100053042; URL, http://www.chictr.org.cn).

Parameter tuning in machine learning based on radiomics biomarkers of lung cancer.

Lung cancer is one of the most common cancers, and early diagnosis and intervention can improve cancer cure rate. To improve predictive performance of radiomics features for lung cancer by tuning the machine learning model parameters. Using a dataset involving 263 cases (125 benign and 138 malignant) acquired from our hospital, each classifier model is trained and tested using 237 and 26 cases, respectively. We initially extract 867 radiomics features of CT images for model development and then test 10 feature selections and 7 models to determine the best method. We further tune the parameter of the final model to reach the best performance. The adjusted final model is then validated using 224 cases acquired from Lung Image Database Consortium (LIDC) dataset (64 benign and 160 malignant) with the same set of selected radiomics features. During model development, the feature selection via concave minimization method show the best performance of area under ROC curve (AUC = 0.765), followed by l0-norm regularization (AUC = 0.741) and Fisher discrimination criterion (AUC = 0.734). Support vector machine (SVM) and random forest (RF) are the top two machine learning algorithms showing the best performance (AUC = 0.765 and 0.734, respectively), using by the default parameter. After parameter tuning, SVM with linear kernel achieves the best performance (AUC = 0.837), whereas the best tuned RF with the number of trees is 510 and yields a slightly lower performance (AUC = 0.775) in 26 test samples data. During model validation, the SVM and RF models yield AUC = 0.78 and 0.77, respectively. Appropriate quantitative radiomics features and accurate parameters can improve the model's performance to predict lung cancer.

A CT image feature space (CTIS) loss for restoration with deep learning-based methods

Objective. Deep learning-based methods have been widely used in medical imaging field such as detection, segmentation and image restoration. For supervised learning methods in CT image restoration, different loss functions will lead to different image qualities which may affect clinical diagnosis. In this paper, to compare commonly used loss functions and give a better alternative, we studied a widely generalizable framework for loss functions which are defined in the feature space extracted by neural networks. Approach. For the purpose of incorporating prior knowledge, a CT image feature space (CTIS) loss was proposed, which learned the feature space from high quality CT images by an autoencoder. In the absence of high-quality CT images, an alternate loss function, random-weight (RaW) loss in the feature space of images (LoFS) was proposed. For RaW-LoFS, the feature space is defined by neural networks with random weights. Main results. In experimental studies, we used post reconstruction deep learning-based methods in the 2016 AAPM low dose CT grand challenge. Compared with perceptual loss that is widely used, our loss functions performed better both quantitatively and qualitatively. In addition, three senior radiologists were invited for subjective assessments between CTIS loss and RaW-LoFS. According to their judgements, the results of CTIS loss achieved better visual quality. Furtherly, by analyzing each channel of CTIS loss, we also proposed partially constrained CTIS loss. Significance. Our loss functions achieved favorable image quality. This framework can be easily adapted to other tasks and fields.

Imaging and Clinical Parameters for Distinction between Infected and Non-Infected Fluid Collections in CT: Prospective Study Using Extended Microbiological Approach.

The aim of this investigation was to evaluate predictive CT imaging features and clinical parameters to distinguish infected from sterile fluid collections. Detection of infectious agents by advanced microbiological analysis was used as the reference standard. From April 2018 to October 2019, all patients undergoing CT-guided drainages were prospectively enrolled, if drainage material volume was at least 5 mL. Univariate analysis revealed attenuation (p = 0.001), entrapped gas (p < 0.001), fat stranding (p < 0.001), wall thickness (p < 0.001) and enhancement (p < 0.001) as imaging biomarkers and procalcitonin (p = 0.003) as clinical predictive parameters for infected fluid collections. On multivariate analysis, attenuation > 10 HU (p = 0.038), presence of entrapped gas (p = 0.027) and wall enhancement (p = 0.028) were independent parameters for distinguishing between infected and non-infected fluids. Gas entrapment had high specificity (93%) but low sensitivity (48%), while wall enhancement had high sensitivity (91%) but low specificity (50%). CT attenuation > 10 HU showed intermediate sensitivity (74%) and specificity (70%). Evaluation of the published proposed scoring systems did not improve diagnostic accuracy over independent predictors in our study. In conclusion, this prospective study confirmed that CT attenuation > 10 HU, entrapped gas and wall enhancement are the key imaging features to distinguish infected from sterile fluid collections on CT.

Multimodal imaging analysis of the cyst like lesion of condyle in temporomandibular joint

Objective: To investigate the imaging features of condylar cystic degeneration of temporomandibular joint (TMJ) by cone-beam CT (CBCT), spiral CT, MRI and radionuclide bone imaging. Methods: From January 2018 to December 2020, thirty-two patients with cyst-like lesions of condylein temporomandibular joint were examined by CBCT, spiral CT, MRI and radionuclide bone imaging at the Department of Oral and Maxillofacial Surgery in General Hospital of Chinese PLA. There were 12 males and 20 females involved, aged from 16 to 65 years with an average age of (33.9±12.5) years. The characteristics of CBCT, spiral CT, MRI and radionuclide bone imaging were analyzed. Condylar cyst like lesions were classified as type A and type B based on the presence or absence of surface bone defects. Condylar cyst like lesions were classified as type Ⅰ(yes) and type Ⅱ(no) according to the accompanying bone marrow edema-like lesions of the condyles. The incidence of condylar bone marrow edema, disc displacement and abnormal bone metabolism were analyzed. Results: A total of 64 joint images of 32 patients were included, including 34 sides with TMJ cyst-like lesion and 6 sides with multiple cyst-like lesions,the total cyst-like lesions were 42. The largest diameter of cyst-like lesion ranged from 1.0 to 12.4 mm, with an average length of (3.7± 1.8) mm. There were 24 cases of type A TMJ cyst like lesion and 10 cases of type B cyst-like lesion. The detection rate of CBCT was 95.2% (40/42) and that of spiral CT was 100% (42/42), there was no significant difference (Calibration Chi-square=0.51, P=0.474). The detection rate of nuclear magnetic resonance was 80.1% (34/42), and the detection rate of cyst-like lesions less than 2 mm was 3/11. In the cyst like lesion side, there were 9 sides with anterior disc displacement with reduction, 20 sides with anterior disc displacement without reduction. In the non-cyst like lesion side, 10 sides with anterior disc displacement with reduction and 6 sides with anterior disc displacement without reduction. There was a significant difference in the displacement of the disc between cyst-like and non-cystic lesion side (χ²=7.80, P=0.005). MRI showed that 6 cases of cystic side[17.6% (6/34)] had bone marrow edema-like lesions (all type A), 1 case of non-cyst like lesions side [3.3% (1/30)] had bone marrow edema-like lesion, there was no significant difference between cystic and non-cystic lesions (Calibration Chi-square=2.04, P=0.153). There was a significant difference between type A and B cystic lesions (Fisher exact probability method, P=0.024). Radionuclide bone imaging showed abnormal bone metabolism in 26 patients in the cyst-like lesion side and 5 patients in the non-cyst like lesion side (χ²=22.82, P<0.001). Conclusions Multi-slice Spiral CT could detect the cyst-like lesion of TMJ condyle in the early stage, which is different from the large joint. And the formation mechanism may vary from the different classifications.

Multimodal imaging analysis of the cyst like lesion of condyle in temporomandibular joint

Objective: To investigate the imaging features of condylar cystic degeneration of temporomandibular joint (TMJ) by cone-beam CT (CBCT), spiral CT, MRI and radionuclide bone imaging. Methods: From January 2018 to December 2020, thirty-two patients with cyst-like lesions of condylein temporomandibular joint were examined by CBCT, spiral CT, MRI and radionuclide bone imaging at the Department of Oral and Maxillofacial Surgery in General Hospital of Chinese PLA. There were 12 males and 20 females involved, aged from 16 to 65 years with an average age of (33.9±12.5) years. The characteristics of CBCT, spiral CT, MRI and radionuclide bone imaging were analyzed. Condylar cyst like lesions were classified as type A and type B based on the presence or absence of surface bone defects. Condylar cyst like lesions were classified as type Ⅰ(yes) and type Ⅱ(no) according to the accompanying bone marrow edema-like lesions of the condyles. The incidence of condylar bone marrow edema, disc displacement and abnormal bone metabolism were analyzed. Results: A total of 64 joint images of 32 patients were included, including 34 sides with TMJ cyst-like lesion and 6 sides with multiple cyst-like lesions,the total cyst-like lesions were 42. The largest diameter of cyst-like lesion ranged from 1.0 to 12.4 mm, with an average length of (3.7±1.8) mm. There were 24 cases of type A TMJ cyst like lesion and 10 cases of type B cyst-like lesion. The detection rate of CBCT was 95.2% (40/42) and that of spiral CT was 100% (42/42), there was no significant difference (Calibration Chi-square=0.51, P=0.474). The detection rate of nuclear magnetic resonance was 80.1% (34/42), and the detection rate of cyst-like lesions less than 2 mm was 3/11. In the cyst like lesion side, there were 9 sides with anterior disc displacement with reduction, 20 sides with anterior disc displacement without reduction. In the non-cyst like lesion side, 10 sides with anterior disc displacement with reduction and 6 sides with anterior disc displacement without reduction. There was a significant difference in the displacement of the disc between cyst-like and non-cystic lesion side (χ²=7.80, P=0.005). MRI showed that 6 cases of cystic side[17.6% (6/34)] had bone marrow edema-like lesions (all type A), 1 case of non-cyst like lesions side [3.3% (1/30)]had bone marrow edema-like lesion, there was no significant difference between cystic and non-cystic lesions (Calibration Chi-square=2.04, P=0.153). There was a significant difference between type A and B cystic lesions (Fisher exact probability method, P= 0.024). Radionuclide bone imaging showed abnormal bone metabolism in 26 patients in the cyst-like lesion side and 5 patients in the non-cyst like lesion side (χ²=22.82, P<0.001). Conclusions Multi-slice Spiral CT could detect the cyst-like lesion of TMJ condyle in the early stage, which is different from the large joint. And the formation mechanism may vary from the different classifications.

Prediction of acute versus chronic osteoporotic vertebral fracture using radiomics-clinical model on CT

PurposeThis paper aims to use radiomics-clinical analysis based on CT imaging to distinguish between acute and chronic osteoporotic vertebral fractures. MethodA total of 147 patients who underwent both dual-energy X-ray absorptiometry (DEXA), CT and MRI of the spine were analyzed retrospectively. The patients were assigned to either a training cohort (n = 103) or a validation cohort (n = 44). The radiomics model and combined nomogram model were established by multivariate logistic regression analysis. The predictive performance was assessed with receiver operating characteristic (ROC) curve. ResultsFourteen radiomic features based on spine CT images were constructed to distinguish acute versus chronic osteoporotic vertebral fractures, and its differentialperformance was good with an area under the curve (AUC) of 0.90 (95% CI, 0.84–0.95) in the training cohort and 0.82 (95% CI, 0.69–0.94) in the validation cohort. Based on the radiomic signature and clinical fracture line feature, a combined nomogram was developed and showed excellent differential ability with highest AUC of 0.93 (95 %CI,0.88–0.98) in the training cohort and 0.86 (95 %CI,0.73–0.98) in the validation cohort, which performed better than the clinical model significantly only. ConclusionsA quantitative nomogram based on clinical fracture line feature and radiomic features of CT images can be used to distinguish acute and chronic osteoporotic vertebral fractures with excellent predictive ability, which can be served as a potential decision support tool to assist clinicians in evaluating the phase of vertebral fractures timely, especially in situation where spine MRI was not available for patient.

Automatic segmentation of high-risk clinical target volume for tandem-and-ovoids brachytherapy patients using an asymmetric dual-path convolutional neural network.

Preimplant diagnostic magnetic resonance imaging is the gold standard for image-guided tandem-and-ovoids (T&O) brachytherapy for cervical cancer. However, high dose rate brachytherapy planning is typically done on postimplant CT-based high-risk clinical target volume (HR-CTVCT ) because the transfer of preimplant Magnetic resonance (MR)-based HR-CTV (HR-CTVMR ) to the postimplant planning CT is difficult due to anatomical changes caused by applicator insertion, vaginal packing, and the filling status of the bladder and rectum. This study aims to train a dual-path convolutional neural network (CNN) for automatic segmentation of HR-CTVCT on postimplant planning CT with guidance from preimplant diagnostic MR. Preimplant T2-weighted MR and postimplant CT images for 65 (48 for training, eight for validation, and nine for testing) patients were retrospectively solicited from our institutional database. MR was aligned to the corresponding CT using rigid registration. HR-CTVCT and HR-CTVMR were manually contoured on CT and MR by an experienced radiation oncologist. All images were then resampled to a spatial resolution of 0.5× 0.5 × 1.25mm. A dual-path 3D asymmetric CNN architecture with two encoding paths was built to extract CT and MR image features. The MR was masked by HR-CTVMR contour while the entire CT volume was included. The network put an asymmetric weighting of 18:6 for CT: MR. Voxel-based dice similarity coefficient (DSCV ), sensitivity, precision, and 95% Hausdorff distance (95-HD) were used to evaluate model performance. Cross-validation was performed to assess model stability. The study cohort was divided into a small tumor group(<20 cc), medium tumor group (20-40cc), and large tumor group(>40cc) based on the HR-CTVCT for model evaluation. Single-path CNN models were trained with the same parameters as those in dual-path models. For this patient cohort, the dual-path CNN model improved each of our objective findings, including DSCV , sensitivity, and precision, with an average improvement of 8%, 7%, and 12%, respectively. The 95-HD was improved by an average of 1.65mm compared to the single-path model with only CT images as input. In addition, the area under the curve for different networks was 0.86 (dual-path with CT and MR) and 0.80 (single-path with CT), respectively. The dual-path CNN model with asymmetric weighting achieved the best performance with DSCV of 0.65 ± 0.03 (0.61-0.70), 0.79 ± 0.02 (0.74-0.85), and 0.75 ± 0.04 (0.68-0.79) for small, medium, and large group. 95-HD were 7.34 (5.35-10.45) mm, 5.48 (3.21-8.43) mm, and 6.21 (5.34-9.32) mm for the three size groups, respectively. An asymmetric CNN model with two encoding paths from preimplant MR (masked by HR-CTVMR ) and postimplant CT images was successfully developed for automatic segmentation of HR-CTVCT for T&O brachytherapy patients.

Gastrointestinal stromal tumours: Preoperative imaging features to predict recurrence after curative resection

PurposeTo identify whether preoperative factors could predict the recurrence after curative resection of gastrointestinal stromal tumours (GISTs) and evaluate the performance of a prediction model using preoperative factors for GIST recurrence compared to a model using preoperative/postoperative factors. MethodThis retrospective study included patients who underwent curative resection and preoperative CT for GIST. CT imaging features as preoperative factors were analysed by two abdominal radiologists. Modified National Institutes of Health scores were accessed as a postoperative factor. Multiple logistic regression analysis was performed to assess the preoperative and postoperative factors in predicting GIST recurrence. Through the logistic regression, two prediction models using preoperative factors only and both preoperative/postoperative factors were constructed, respectively. The internal validation of the prediction models was performed using bootstrapping sampling. ResultsData in 113 patients were evaluated. Among them, 14 patients had recurrence. The multiple logistic regression analysis demonstrated that non-gastric location (odds ratio [OR] = 5.12, p = 0.029), ill-defined margin (OR = 4.93, p = 0.023), and prominent vessels around tumour (OR = 6.78, p = 0.007) were significant factors. The prediction models using these preoperative factors and adding a postoperative factor showed an area under the receiver operating characteristic curve of 0.863 and 0.897, respectively, which were not statistically different. The bootstrapping sampling showed the two models were valid. ConclusionThe prediction model derived from non-gastric location, ill-defined margin, and prominent vessels around tumour can be used preoperatively to estimate the risk of recurrence after resection of GIST.

Analysis of the value of enhanced CT combined with texture analysis in the differential diagnosis of pulmonary sclerosing pneumocytoma and atypical peripheral lung cancer: a feasibility study

BackgroundAs a rare benign lung tumour, pulmonary sclerosing pneumocytoma (PSP) is often misdiagnosed as atypical peripheral lung cancer (APLC) on routine imaging examinations. This study explored the value of enhanced CT combined with texture analysis to differentiate between PSP and APLC.MethodsForty-eight patients with PSP and fifty patients with APLC were retrospectively enrolled. The CT image features of the two groups of lesions were analysed, and MaZda software was used to evaluate the texture of CT venous phase thin-layer images. Independent sample t-test, Mann–Whitney U tests or χ2 tests were used to compare between groups. The intra-class correlation coefficient (ICC) was used to analyse the consistency of the selected texture parameters. Spearman correlation analysis was used to evaluate the differences in texture parameters between the two groups. Based on the statistically significant CT image features and CT texture parameters, the independent influencing factors between PSP and APLC were analysed by multivariate logistic regression. Extremely randomized trees (ERT) was used as the classifier to build models, and the models were evaluated by the five-fold cross-validation method.ResultsLogistic regression analysis based on CT image features showed that calcification and arterial phase CT values were independent factors for distinguishing PSP from APLC. The results of logistic regression analysis based on CT texture parameters showed that WavEnHL_s-1 and Perc.01% were independent influencing factors to distinguish the two. Compared with the single-factor model (models A and B), the classification accuracy of the model based on image features combined with texture parameters was 0.84 ± 0.04, the AUC was 0.84 ± 0.03, and the sensitivity and specificity were 0.82 ± 0.13 and 0.87 ± 0.12, respectively.ConclusionEnhanced CT combined with texture analysis showed good diagnostic value for distinguishing PSP and APLC, which may contribute to clinical decision-making and prognosis evaluation.

Identifying Prognostic Markers From Clinical, Radiomics, and Deep Learning Imaging Features for Gastric Cancer Survival Prediction.

BackgroundGastric cancer is one of the leading causes of cancer death in the world. Improving gastric cancer survival prediction can enhance patient prognostication and treatment planning.MethodsIn this study, we performed gastric cancer survival prediction using machine learning and multi-modal data of 1061 patients, including 743 for model learning and 318 independent patients for evaluation. A Cox proportional-hazard model was trained to integrate clinical variables and CT imaging features (extracted by radiomics and deep learning) for overall and progression-free survival prediction. We further analyzed the prediction effects of clinical, radiomics, and deep learning features. Concordance index (c-index) was used as the model performance metric, and the predictive effects of multi-modal features were measured by hazard ratios (HRs) at pre- and post-operative settings.ResultsAmong 318 patients in the independent testing group, the hazard predicted by Cox from multi-modal features is associated with their survival. The highest c-index was 0.783 (95% CI, 0.782-0.783) and 0.770 (95% CI, 0.769-0.771) for overall and progression-free survival prediction, respectively. The post-operative variables are significantly (p<0.001) more predictive than the pre-operative variables. Pathological tumor stage (HR=1.336 [overall survival]/1.768 [progression-free survival], p<0.005), pathological lymph node stage (HR=1.665/1.433, p<0.005), carcinoembryonic antigen (CEA) (HR=1.632/1.522, p=0.02), chemotherapy treatment (HR=0.254/0.287, p<0.005), radiomics signature [HR=1.540/1.310, p<0.005], and deep learning signature [HR=1.950/1.420, p<0.005]) are significant survival predictors.ConclusionOur study showed that CT radiomics and deep learning imaging features are significant pre-operative predictors, providing additional prognostic information to the pathological staging markers. Lower CEA levels and chemotherapy treatments also increase survival chances. These findings can enhance gastric cancer patient prognostication and inform treatment planning.

Reproducibility of radiomic features in CT images of NSCLC patients: an integrative analysis on the impact of acquisition and reconstruction parameters

BackgroundWe investigated to what extent tube voltage, scanner model, and reconstruction algorithm affect radiomic feature reproducibility in a single-institution retrospective database of computed tomography images of non-small-cell lung cancer patients.MethodsThis study was approved by the Institutional Review Board (UID 2412). Images of 103 patients were considered, being acquired on either among two scanners, at 100 or 120 kVp. For each patient, images were reconstructed with six iterative blending levels, and 1414 features were extracted from each reconstruction. At univariate analysis, Wilcoxon-Mann-Whitney test was applied to evaluate feature differences within scanners and voltages, whereas the impact of the reconstruction was established with the overall concordance correlation coefficient (OCCC). A multivariable mixed model was also applied to investigate the independent contribution of each acquisition/reconstruction parameter. Univariate and multivariable analyses were combined to analyse feature behaviour.ResultsScanner model and voltage did not affect features significantly. The reconstruction blending level showed a significant impact at both univariate analysis (154/1414 features yielding an OCCC < 0.85) and multivariable analysis, with most features (1042/1414) revealing a systematic trend with the blending level (multiple comparisons adjusted p < 0.05). Reproducibility increased in association to image processing with smooth filters, nonetheless specific investigation in relation to clinical endpoints should be performed to ensure that textural information is not removed.ConclusionsCombining univariate and multivariable models is allowed to identify features for which corrections may be applied to reduce the trend with the algorithm and increase reproducibility. Subsequent clustering may be applied to eliminate residual redundancy.