CT Features Research Articles

A machine learning-based PET/CT model for automatic diagnosis of early-stage lung cancer.

The aim of this study was to develop a machine learning-based automatic analysis method for the diagnosis of early-stage lung cancer based on positron emission tomography/computed tomography (PET/CT) data. A retrospective cohort study was conducted using PET/CT data from 187 cases of non-small cell lung cancer (NSCLC) and 190 benign pulmonary nodules. Twelve PET and CT features were used to train a diagnosis model. The performance of the machine learning-based PET/CT model was tested and validated in two separate cohorts comprising 462 and 229 cases, respectively. The standardized uptake value (SUV) was identified as an important biochemical factor for the early stage of lung cancer in this model. The PET/CT diagnosis model had a sensitivity and area under the curve (AUC) of 86.5% and 0.89, respectively. The testing group comprising 462 cases showed a sensitivity and AUC of 85.7% and 0.87, respectively, while the validation group comprising 229 cases showed a sensitivity and AUC of 88.4% and 0.91, respectively. Additionally, the proposed model improved the clinical discrimination ability for solid pulmonary nodules (SPNs) in the early stage significantly. The feature data collected from PET/CT scans can be analyzed automatically using machine learning techniques. The results of this study demonstrated that the proposed model can significantly improve the accuracy and positive predictive value (PPV) of SPNs at the early stage. Furthermore, this algorithm can be optimized into a robotic and less biased PET/CT automatic diagnosis system.

Differentiation of pulmonary solid nodules attached to the pleura detected by thin-section CT

BackgroundPulmonary solid pleura-attached nodules (SPANs) are not very commonly detected and thus not well studied and understood. This study aimed to identify the clinical and CT characteristics for differentiating benign and malignant SPANs.ResultsFrom January 2017 to March 2023, a total of 295 patients with 300 SPANs (128 benign and 172 malignant) were retrospectively enrolled. Between benign and malignant SPANs, there were significant differences in patients’ age, smoking history, clinical symptoms, CT features, nodule-pleura interface, adjacent pleural change, peripheral concomitant lesions, and lymph node enlargement. Multivariate analysis revealed that smoking history (odds ratio [OR], 2.016; 95% confidence interval [CI], 1.037–3.919; p = 0.039), abutting the mediastinal pleura (OR, 3.325; 95% CI, 1.235–8.949; p = 0.017), nodule diameter (> 15.6 mm) (OR, 2.266; 95% CI, 1.161–4.423; p = 0.016), lobulation (OR, 8.922; 95% CI, 4.567–17.431; p < 0.001), narrow basement to pleura (OR, 6.035; 95% CI, 2.847–12.795; p < 0.001), and simultaneous hilar and mediastinal lymph nodule enlargement (OR, 4.971; 95% CI, 1.526–16.198; p = 0.008) were independent predictors of malignant SPANs, and the area under the curve (AUC) of this model was 0.890 (sensitivity, 82.0%, specificity, 77.3%) (p < 0.001).ConclusionIn patients with a smoking history, SPANs abutting the mediastinal pleura, having larger size (> 15.6 mm in diameter), lobulation, narrow basement, or simultaneous hilar and mediastinal lymph nodule enlargement are more likely to be malignant.Critical relevance statementThe benign and malignant SPANs have significant differences in clinical and CT features. Understanding the differences between benign and malignant SPANs is helpful for selecting the high-risk ones and avoiding unnecessary surgical resection.Key points• The solid pleura-attached nodules (SPANs) are closely related to the pleura.• Relationship between nodule and pleura and pleural changes are important for differentiating SPANs.• Benign SPANs frequently have broad pleural thickening or embed in thickened pleura.• Smoking history and lesions abutting the mediastinal pleura are indicators of malignant SPANs.• Malignant SPANs usually have larger diameters, lobulation signs, narrow basements, and lymphadenopathy.Graphical

A Method to Evaluate the Reduction Difficulty of Atlantoaxial Dislocation Based on Computed Tomography Quantitative Analysis

To evaluate the difficulty of reduction of congenital atlantoaxial dislocation with or without os odontoideum or basilar invagination based on computed tomography (CT) quantitative analysis. From March 2018 to December 2022, the CT features of 108 patients with atlantoaxial dislocation with or without os odontoideum or basilar invagination were analyzed. Quantitative scores were defined according to imaging features, including sloping of the lateral mass; osteophyte between the lateral mass joint; ball-and-socket deformity of the lateral mass joint; vertical interlocking of the lateral mass joint; callus between the lateral mass joint; and atlanto-odontoid joint hyperplasia, blocking, or fusion. Grades were calculated according to the sum of points of the atlanto-odontoid joint and lateral mass joints, as follows: I, 0-1 points; Ⅱ, 2-3 points; Ⅲ, 4-6 points; IV, 7-10 points. After 1 week of bidirectional cervical traction, CT scans were performed, and atlantodens interval and vertical distance from dens to Chamberlain line were measured. The vertical reduction rate, horizontal reduction rate, and overall reduction rate of atlantoaxial dislocation were calculated. The vertical distance from dens to Chamberlain line values after traction were significantly reduced compared with before traction, including grades I, II, III, and IV. The overall reduction rates were 85.1% ± 11.8%, 65.8% ± 8.3%, 45.0% ± 8.5%, and 38.4% ± 13.0% respectively, after 1 week of bidirectional cervical traction. The CT quantitative score system is an effective noninvasive evaluation to judge the reduction difficulty of atlantoaxial dislocation with or without os odontoideum or basilar invagination.

Radiomics preoperative-Fistula Risk Score (RAD-FRS) for pancreatoduodenectomy: development and external validation.

Accurately predicting the risk of clinically relevant postoperative pancreatic fistula after pancreatoduodenectomy before surgery may assist surgeons in making more informed treatment decisions and improved patient counselling. The aim was to evaluate the predictive accuracy of a radiomics-based preoperative-Fistula Risk Score (RAD-FRS) for clinically relevant postoperative pancreatic fistula. Radiomic features were derived from preoperative CT scans from adult patients after pancreatoduodenectomy at a single centre in the Netherlands (Amsterdam, 2013-2018) to develop the radiomics-based preoperative-Fistula Risk Score. Extracted radiomic features were analysed with four machine learning classifiers. The model was externally validated in a single centre in Italy (Verona, 2020-2021). The radiomics-based preoperative-Fistula Risk Score was compared with the Fistula Risk Score and the updated alternative Fistula Risk Score. Overall, 359 patients underwent a pancreatoduodenectomy, of whom 89 (25 per cent) developed a clinically relevant postoperative pancreatic fistula. The radiomics-based preoperative-Fistula Risk Score model was developed using CT scans of 118 patients, of which three radiomic features were included in the random forest model, and externally validated in 57 patients. The model performed well with an area under the curve of 0.90 (95 per cent c.i. 0.71 to 0.99) and 0.81 (95 per cent c.i. 0.69 to 0.92) in the Amsterdam test set and Verona data set respectively. The radiomics-based preoperative-Fistula Risk Score performed similarly to the Fistula Risk Score (area under the curve 0.79) and updated alternative Fistula Risk Score (area under the curve 0.79). The radiomics-based preoperative-Fistula Risk Score, which uses only preoperative CT features, is a new and promising radiomics-based score that has the potential to be integrated with hospital CT report systems and improve patient counselling before surgery. The model with underlying code is readily available via www.pancreascalculator.com and www.github.com/PHAIR-Consortium/POPF-predictor.

Cone-beam CT imaging features of common cystic lesions associated with the impacted mandibular third molar

Objective: To analyze the imaging features of cone-beam CT (CBCT) of ameloblastoma (AB), odontogenic keratocyst (OKC) and dentigerous cysts (DC) associated with the mandibular impacted third molars,so as to provide useful information for differential diagnosis of these lesions. Methods: The patients who were with complete clinical data, pathological diagnosis and CBCT images from AB, OKC and DC around the mandibular impacted third molars were collected in Peking University Hospital of Stomatology from August 2016 to December 2021. A total of 109 patients (14 were diagnosed as AB, 23 were diagnosed as OKC and the others were diagnosed as dental cysts) were collected, including 73 males and 36 females. The age ranged from 11 to 70. The analyzed imaging features included location and internal density of the lesions, bone expansion, root resorption of adjacent teeth and types of the impacted teeth. The Chi square test was used to compare the gender of different lesions, and the Fisher's exact test was used to compare imaging features of lesions. When P<0.05, there was a significant difference among the three. Logistic regression analysis was performed to determine the imaging features that significantly contribute to correct imaging diagnosis. Corresponding P-values were calculated for all factors from multivariate models. Results: In the 23 cases of OKC, no special location was observed for the center of the lesion, heterogeneous high-density were seen in 21.7% of the cases, 56.5% of the cases had no significant bone expansion and the impacted teeth were not specially oriented. Among the 14 AB, 7 cases (7/14) were mainly located in the ramus of the mandible, and all cases (14/14) had buccal/lingual expansion of the jaw, 8 cases (8/14) presented root resorption of the adjacent teeth, and mesial impacted mandibular third molar were seen in 6 cases (6/14). Among the 72 DC, 88.9% (64/72) of the cases were mainly limited to the crown of the impacted third molar, 72.2% (52/72) of the cases had no obvious bone expansion, inverted impacted teeth were shown in 56.9% (41/72) of the cases. There was a significant difference among the three groups (χ2=7.30, P=0.026) in gender. AB and odontogenic cyst were more common in men than in women, while the incidence of OKC was roughly equal between men and women.There were significant differences in the location (P<0.001), internal density (P=0.001) of the lesions, bone expansion (P<0.001) and types of the impacted teeth (P<0.001), while no statistical difference was found for root resorption of adjacent teeth (P=0.153). Logistics regression analysis showed that the location of the lesion, internal density, bone expansion, root resorption of adjacent teeth and the types of impacted teeth had significant effects on the accurate diagnosis of the three kinds of lesions. Conclusions: Location, internal density, bone expansion and types of the impacted teeth played an important role in the correct imaging diagnosis. Further analysis indicates that when the classification of impacted teeth and the location of lesions are considered synchronously, DC can be differentiated from AB and OKC.

Unenhanced computed tomography radiomics help detect endoleaks after endovascular repair of abdominal aortic aneurysm

ObjectivesTo explore the feasibility of unenhanced CT images for endoleak detection of abdominal aortic aneurysm (AAA) after endovascular repair (EVAR).MethodsPatients who visited our hospital after EVAR from July 2014 to September 2021 were retrospectively collected. Two radiologists evaluated the presence or absence of endoleaks using the combination of contrast-enhanced and unenhanced CT as the referenced standard. After segmenting the aneurysm sac of the unenhanced CT, the radiomic features were automatically extracted from the region of interest. Histogram features of patients with and without endoleak were statistically analyzed to explore the differences between the two groups. Twelve common machine learning (ML) models based on radiomic features were constructed to evaluate the performance of endoleak detection with unenhanced CT images.ResultsThe study included 216 patients (69 ± 8 years; 191 men) with AAA, including 64 patients with endoleaks. A total of 1955 radiomic features of unenhanced CT were extracted. Compared with patients without endoleak, the aneurysm sac outside the stent of patients with endoleak had higher CT attenuation (41.7 vs. 33.6, p < 0.001) with smaller dispersion (51.5 vs. 58.8, p < 0.001). The average area under the curve (AUC) of the ML models constructed with unenhanced CT radiomics was 0.86 ± 0.05, the accuracy was 81% ± 4, the sensitivity was 88% ± 10, and the specificity was 78% ± 5. When fixing the sensitivity to > 90% (92% ± 2), the models retained specificity at 72% ± 10.ConclusionsUnenhanced CT features exhibit significant differences between patients with and without endoleak and can help detect endoleaks in AAA after EVAR with high sensitivity.Clinical relevance statementUnenhanced CT radiomics can help provide an alternative method of endoleak detection in patients who have adverse reactions to contrast media. This study further exploits the value of unenhanced CT examinations in the clinical management and surveillance of postoperative abdominal aortic aneurysm.Key Points• Unenhanced CT features of the aneurysm sac outside the stent exhibit significant differences between patients with and without endoleak. The endoleak group showed higher unenhanced CT attenuation (41.7 vs 33.6, p < .001) with smaller dispersion (51.5 vs 58.8, p < .001) than the nonendoleak group.• Unenhanced CT radiomics can help detect endoleaks after intervention. The average area under the curve (AUC) of twelve common machine learning models constructed with unenhanced CT radiomics was 0.86 ± 0.05, the average accuracy was 81% ± 4.• When fixing the sensitivity to > 90% (92% ± 2), the machine learning models retained average specificity at 72% ± 10.

Short-term mortality prediction using a combination of clinical and CT features: Refining the prognosis of critically ill patients in shock

PurposeTo assess the predictive value of combining CT and clinical findings for predicting 10-day mortality in critically ill patients in shock. Materials and methodsFrom January 1, 2018, to December 31, 2021, 289 consecutives critically ill patients in shock who underwent a contrast enhanced CT were included. Variables at the time of the CT were retrospectively extracted from medical charts. CT examinations were blindly analyzed by two independent radiologists. Multivariable analysis was performed, combining clinical and CT features. A simple survival score for 10-day mortality prediction was built and validated in a further independent external cohort of 70 patients. Results10-day mortality rate was 135/289 (47%) in the study sample. At multivariate analysis, catecholamine infusion (OR = 2.11; 95%CI [1.21–4.18], P = 0.011), lactates level > 5 mmol/l (OR = 3.54; 95%CI [1.94–6.54], P < 0.001); total bilirubin > 50 mg/l (OR = 1.79 CI 95% [1.03–3.13], P = 0.039); small bowel dilation (OR = 1.82; 95%CI [1.01–3.32], P = 0.047); diffuse kidney infarction (OR = 2.76; 95%CI [1.26–6.37], P = 0.013) and superior mesentery artery < 5 mm (OR = 1.96; 95%CI [1.10–3.49], P = 0.021) were associated with 10-days mortality. The AUC of the combined model was 0.79; 95%CI [0.74–0.85] in the study sample and 0.87; 95%CI [0.71–0.91] in the validation cohort. ConclusionThe combination of CT imaging features and clinical data should emerge as a novel approach to predict short-term mortality in critically ill patients in shock.

Automatic statistical diagnosis of COVID-19 based on multi-modal CT feature extraction

Background and purposeComputed tomography (CT) is highly sensitive to lung-related abnormalities as a non-invasive method and has become an essential tool for screening and diagnosing Coronavirus disease 2019 (COVID-19). To reduce the stress of work for physicians and speed up diagnosis, we propose a novel automatic diagnosis pipeline for COVID-19 based on high-dimensional radiomic features extracted from multimodal CT scans (multi-geometric and multiscale). Materials and methodsThere are 746 CT scans involved in this study, where 349 CT scans are COVID-19 positive and 397 CT scans are COVID-19 negative. All of them are from the public dataset. We first construct a transfer learning-based auto-segmentation model with a morphological post-processing block to improve the lung region segmentation. Then the radiomics feature extraction is guided by the proposed multi-modal CT scans strategy. In addition, our automatic diagnosis pipeline is driven by a well-designed loss function. We also explain the diagnosis capability from the related theory of linear subspace spanned by multi-modal radiomics features. ResultsUnder the 10-fold cross-validation strategy, our approach can achieve an improvement in diagnostic performance of 5. 77%, 7. 78%, 7. 74%, 7. 78%, 7. 45% compared to the radiomic features extracted from the original CT scans, and diagnosis performance is promoted to 91.53%, 86.46%, 86.47%, 86.46%, 86.95% in terms of AUC, Acc, F1, Recall and Precision in public datasets. ConclusionsWe demonstrate a statistically significant improvement of the proposed statistical learning method compared to the state-of-the-art machine learning-based diagnosis approaches. Thanks to theoretical support and excellent diagnostic performance, our method can be deployed in clinical auxiliary diagnosis, releasing the overstretched medical resources.

Differentiation between Chondrosarcoma and Synovial Chondromatosis of the Temporomandibular Joint Using CT and MR Imaging.

Chondrosarcoma and synovial chondromatosis of the temporomandibular joint share overlapping clinical and histopathologic features. We aimed to identify CT and MR imaging features to differentiate chondrosarcoma from synovial chondromatosis of the temporomandibular joint. The CT and MR images of 12 and 35 patients with histopathologically confirmed chondrosarcoma and synovial chondromatosis of the temporomandibular joint, respectively, were retrospectively reviewed. Imaging features including lesion size, center, enhancement, destruction/sclerosis of surrounding bone, infiltration into the tendon of the lateral pterygoid muscle, calcification, periosteal reaction, and osteophyte formation were assessed. A comparison between chondrosarcoma and synovial chondromatosis was performed with a Student t test for quantitative variables and the Fisher exact test or linear-by-linear association test for qualitative variables. Receiver operating characteristic analysis was performed to determine the diagnostic performance for differentiation of chondrosarcoma and synovial chondromatosis based on a composite score obtained by assigning 1 point for each of 9 imaging features. High-risk imaging features for chondrosarcoma were the following: lesion centered on the mandibular condyle, destruction of the mandibular condyle, no destruction/sclerosis of the articular eminence/glenoid fossa, infiltration into the tendon of the lateral pterygoid muscle, absent or stippled calcification, periosteal reaction, internal enhancement, and size of ≥30.5 mm. The best cutoff value to discriminate chondrosarcoma from synovial chondromatosis was the presence of any 4 of these high-risk imaging features, with an area under the curve of 0.986 and an accuracy of 95.8%. CT and MR imaging features can distinguish chondrosarcoma from synovial chondromatosis of the temporomandibular joint with improved diagnostic performance when a subcombination of 9 imaging features is used.

Association of blood total immunoglobulin E and eosinophils with radiological features of bronchiectasis

BackgroundOur study aimed to investigate whether serum total IgE and blood eosinophils were associated with radiological features of bronchiectasis in a Chinese cohort.MethodsWe retrospectively enrolled bronchiectasis patients who visited Peking University Third Hospital from Jan 1st, 2012 to Oct 7th, 2021. The clinical, laboratory and chest CT characteristics were analyzed in association with serum total IgE level and blood eosinophil count.ResultsA total of 125 bronchiectasis patients were enrolled, with 50.4% (63/125) female, and a mean age of 62.4 ± 14.1 years. The median serum total IgE level and blood eosinophil count were 47.7 (19.8, 123.0) KU/L and 140 (90, 230) cells/µl, respectively. In patients with a higher than normal (normal range, 0–60 KU/L) total IgE (43.2%, n = 54), more lobes were involved [4 (3, 5) vs. 3 (2, 4), p = 0.008], and mucus plugs were more common (25.9% vs. 9.9%, p =0.017) on HRCT, as compared to those with a normal level of total IgE. The higher IgE group was more likely to have bilateral involvement (p = 0.059), and had numerically higher Smith and Bhalla scores, but the differences were not statistically significant. In patients with an eosinophil count ≥ 150 cells/µl (49.6%, n = 62), the number of lobes involved was greater [4 (3, 5) vs. 3 (2, 4), p = 0.015], and the Smith and Bhalla scores were higher [9 (5, 12) vs. 6 (3, 9), p = 0.009, 7 (5, 11) vs. 5 (3, 9), p = 0.036]. The Smith score was correlated positively with the eosinophil count (r = 0.207, p = 0.020). Fractional exhaled nitric oxide (FeNO) was correlated with total IgE (r = 0.404, p = 0.001) and eosinophil count (r = 0.310, p = 0.014).ConclusionsOur study demonstrated that serum total IgE and the blood eosinophil count were associated with the radiological extent and severity of bronchiectasis, necessitating further investigation on the role of T2 inflammation in structural abnormalities of this heterogeneous disease.

Analysis of tissue lipidomics and computed tomography pulmonary fat attenuation volume (CTPFAV ) in idiopathic pulmonary fibrosis.

There is increasing interest in the role of lipids in processes that modulate lung fibrosis with evidence of lipid deposition in idiopathic pulmonary fibrosis (IPF) histological specimens. The aim of this study was to identify measurable markers of pulmonary lipid that may have utility as IPF biomarkers. IPF and control lung biopsy specimens were analysed using a unbiased lipidomic approach. Pulmonary fat attenuation volume (PFAV) was assessed on chest CT images (CTPFAV ) with 3D semi-automated lung density software. Aerated lung was semi-automatically segmented and CTPFAV calculated using a Hounsfield-unit (-40 to -200HU) threshold range expressed as a percentage of total lung volume. CTPFAV was compared to pulmonary function, serum lipids and qualitative CT fibrosis scores. There was a significant increase in total lipid content on histological analysis of IPF lung tissue (23.16 nmol/mg) compared to controls (18.66 mol/mg, p = 0.0317). The median CTPFAV in IPF was higher than controls (1.34% vs. 0.72%, p < 0.001) and CTPFAV correlated significantly with DLCO% predicted (R2 = 0.356, p < 0.0001) and FVC% predicted (R2 = 0.407, p < 0.0001) in patients with IPF. CTPFAV correlated with CT features of fibrosis; higher CTPFAV was associated with >10% reticulation (1.6% vs. 0.94%, p = 0.0017) and >10% honeycombing (1.87% vs. 1.12%, p = 0.0003). CTPFAV showed no correlation with serum lipids. CTPFAV is an easily quantifiable non-invasive measure of pulmonary lipids. In this pilot study, CTPFAV correlates with pulmonary function and radiological features of IPF and could function as a potential biomarker for IPF disease severity assessment.

Evaluation of calcification distribution by CT-based textural analysis for discrimination of immature teratoma

BackgroundMature and immature teratomas are differentiated based on tumor markers and calcification or fat distribution. However, no study has objectively quantified the differences in calcification and fat distributions between these tumors. This study aimed to evaluate the diagnostic potential of CT-based textural analysis in differentiating between mature and immature teratomas in patients aged < 20 years.Materials and methodsThirty-two patients with pathologically proven mature cystic (n = 28) and immature teratomas (n = 4) underwent transabdominal ultrasound and/or abdominal and pelvic CT before surgery. The diagnostic performance of CT for assessing imaging features, including subjective manual measurement and objective textural analysis of fat and calcification distributions in the tumors, was evaluated by two experienced readers. The histopathological results were used as the gold standard. The Mann–Whitney U test was used for statistical analysis.ResultsWe evaluated 32 patients (mean age, 14.5 years; age range, 6–19 years). The mean maximum diameter and number of calcifications of immature teratomas were significantly larger than those of mature cystic teratomas (p < 0.01). The mean number of fats of immature teratomas was significantly larger than that of mature cystic teratomas (p < 0.01); however, no significant difference in the maximum diameter of fats was observed. CT textural features for calcification distribution in the tumors showed that mature cystic teratomas had higher homogeneity and energy than immature teratomas. However, immature teratomas showed higher correlation, entropy, and dissimilarity than mature cystic teratomas among features derived from the gray-level co-occurrence matrix (GLCM) (p < 0.05). No significant differences were observed in the CT features of fats derived from GLCM.ConclusionOur results demonstrate that calcification distribution on CT is a potential diagnostic biomarker to discriminate mature from immature teratomas, thus enabling optimal therapeutic selection for patients aged < 20 years.

Duodenal neuroendocrine neoplasms on enhanced CT: establishing a diagnostic model with duodenal gastrointestinal stromal tumors in the non-ampullary area and analyzing the value of predicting prognosis

ObjectiveTo identify CT features and establish a diagnostic model for distinguishing non-ampullary duodenal neuroendocrine neoplasms (dNENs) from non-ampullary duodenal gastrointestinal stromal tumors (dGISTs) and to analyze overall survival outcomes of all dNENs patients.Materials and methodsThis retrospective study included 98 patients with pathologically confirmed dNENs (n = 44) and dGISTs (n = 54). Clinical data and CT characteristics were collected. Univariate analyses and binary logistic regression analyses were performed to identify independent factors and establish a diagnostic model between non-ampullary dNENs (n = 22) and dGISTs (n = 54). The ROC curve was created to determine diagnostic ability. Cox proportional hazards models were created and Kaplan–Meier survival analyses were performed for survival analysis of dNENs (n = 44).ResultsThree CT features were identified as independent predictors of non-ampullary dNENs, including intraluminal growth pattern (OR 0.450; 95% CI 0.206–0.983), absence of intratumoral vessels (OR 0.207; 95% CI 0.053–0.807) and unenhanced lesion > 40.76 HU (OR 5.720; 95% CI 1.575–20.774). The AUC was 0.866 (95% CI 0.765–0.968), with a sensitivity of 90.91% (95% CI 70.8–98.9%), specificity of 77.78% (95% CI 64.4–88.0%), and total accuracy rate of 81.58%. Lymph node metastases (HR: 21.60), obstructive biliary and/or pancreatic duct dilation (HR: 5.82) and portal lesion enhancement ≤ 99.79 HU (HR: 3.02) were independent prognostic factors related to poor outcomes.ConclusionWe established a diagnostic model to differentiate non-ampullary dNENs from dGISTs. Besides, we found that imaging features on enhanced CT can predict OS of patients with dNENs.

Prediction of postoperative recurrence in resectable pancreatic body/tail adenocarcinoma: a novel risk stratification approach using a CT-based nomogram.

To identify prognostic CT features that predict recurrence in patients with resectable pancreatic body/tail adenocarcinoma (PBTA) and construct a CT-based nomogram for preoperative risk stratification. A total of 258 patients with resectable PBTA who underwent upfront surgery were retrospectively enrolled (development cohort, n = 172; validation cohort, n = 86), and their clinical and CT features were analyzed. Stepwise Cox proportional hazard analysis was performed to identify prognostic features and construct a predictive nomogram for recurrence-free survival (RFS). The prognostic performance of the CT-based nomogram was validated and compared to the 8th American Joint Committee on Cancer (AJCC) pathological staging system. In the development cohort, the following five CT features for predicting recurrence were identified to construct the nomogram: tumor density in the venous phase, tumor necrosis, adjacent organ invasion, splenic vein invasion, and superior mesenteric vein/portal vein abutment. In the validation cohort, the CT-based nomogram showed a concordance index of 0.65 (95% confidence interval: 0.58-0.73), which was higher than the 8th AJCC staging system. The area under the curves of the nomogram for predicting recurrence at 0.5, 1, and 2years were 0.66, 0.71, and 0.72, respectively. Patients were categorized into high- and low-risk groups with 1-year recurrence probabilities of 0.73 and 0.43, respectively. The proposed nomogram provided accurate recurrence risk stratification for patients with resectable PBTA in a preoperative setting and may be used to facilitate clinical decision-making. The proposed CT-based nomogram, based on easily available CT features, may serve as an effective and convenient tool for stratifying further the recurrence risk of patients with pancreatic body/tail adenocarcinoma. • The CT-based nomogram, incorporating five commonly used CT features, successfully preoperatively stratified patients with resectable PBTA into distinct prognosis groups. • Tumor density in the venous phase, tumor necrosis, splenic vein invasion, adjacent organ invasion, and superior mesenteric vein/portal vein abutment were associated with RFS in patients with resectable PBTA. • The CT-based nomogram exhibited better predictive performance for recurrence than the 8th AJCC staging system.

A CT-based novel model to predict pathological complete response of locally advanced esophageal squamous cell carcinoma to neoadjuvant PD-1 blockade in combination with chemotherapy

PurposeTo develop a novel CT-based model to predict pathological complete response (pCR) of locally advanced esophageal squamous cell carcinoma (ESCC) to neoadjuvant PD-1 blockade in combination with chemotherapy. Methods117 consecutive patients with locally advanced ESCC were stratified into training cohort (n = 82) and validation cohort (n = 35). All patients underwent non-contrast and contrast-enhanced thoracic and upper abdominal CT before neoadjuvant PD-1 blockade in combination with chemotherapy (CTpre), and after two cycles of the therapy before esophagectomy (CTpost), respectively. Univariate analyses and binary logistic regression analyses of ESCC quantitative and qualitative CT features were performed to determine independent predictors of pCR. Prediction performance of the model developed with independent predictors from training cohort was evaluated by receiver operating characteristic (ROC) analysis, and validated by Kappa test in validation cohort. ResultsIn training cohort, the difference in CT attenuation between tumor and background normal esophageal wall obtained from CTpre (ΔTNpre), tumoral increased CT attenuation after contrast-enhanced scan from CTpost images (ΔTpost) and gross tumor volume (GTV) from CTpre were independent predictors of pCR (odds ratio = 1.128 (95% confidence interval (CI): 0.997–1.277), 1.113 (95%CI: 0.965–1.239) and 1.133 (95%CI: 1.043–1.231), respectively, all P-values < 0.05). Logistic regression model equation (0.121 × ΔTNpre + 0.107 × ΔTpost + 0.125 × GTV − 9.856) to predict pCR showed the best performance with an area under the ROC of 0.876, compared with each independent predictor. The good performance was confirmed by the Kappa test (K-value = 0.796) in validation cohort. ConclusionsThis novel model can be reliable to predict pCR to neoadjuvant PD-1 blockade in combination with chemotherapy in locally advanced ESCC.

Mining the contribution of intensive care clinical course to outcome after traumatic brain injury

Existing methods to characterise the evolving condition of traumatic brain injury (TBI) patients in the intensive care unit (ICU) do not capture the context necessary for individualising treatment. Here, we integrate all heterogenous data stored in medical records (1166 pre-ICU and ICU variables) to model the individualised contribution of clinical course to 6-month functional outcome on the Glasgow Outcome Scale -Extended (GOSE). On a prospective cohort (n = 1550, 65 centres) of TBI patients, we train recurrent neural network models to map a token-embedded time series representation of all variables (including missing values) to an ordinal GOSE prognosis every 2 h. The full range of variables explains up to 52% (95% CI: 50–54%) of the ordinal variance in functional outcome. Up to 91% (95% CI: 90–91%) of this explanation is derived from pre-ICU and admission information (i.e., static variables). Information collected in the ICU (i.e., dynamic variables) increases explanation (by up to 5% [95% CI: 4–6%]), though not enough to counter poorer overall performance in longer-stay (>5.75 days) patients. Highest-contributing variables include physician-based prognoses, CT features, and markers of neurological function. Whilst static information currently accounts for the majority of functional outcome explanation after TBI, data-driven analysis highlights investigative avenues to improve the dynamic characterisation of longer-stay patients. Moreover, our modelling strategy proves useful for converting large patient records into interpretable time series with missing data integration and minimal processing.

Baseline whole-lung CT features deriving from deep learning and radiomics: prediction of benign and malignant pulmonary ground-glass nodules.

To develop and validate the model for predicting benign and malignant ground-glass nodules (GGNs) based on the whole-lung baseline CT features deriving from deep learning and radiomics. This retrospective study included 385 GGNs from 3 hospitals, confirmed by pathology. We used 239 GGNs from Hospital 1 as the training and internal validation set; 115 and 31 GGNs from Hospital 2 and Hospital 3 as the external test sets 1 and 2, respectively. An additional 32 stable GGNs from Hospital 3 with more than five years of follow-up were used as the external test set 3. We evaluated clinical and morphological features of GGNs at baseline chest CT and extracted the whole-lung radiomics features simultaneously. Besides, baseline whole-lung CT image features are further assisted and extracted using the convolutional neural network. We used the back-propagation neural network to construct five prediction models based on different collocations of the features used for training. The area under the receiver operator characteristic curve (AUC) was used to compare the prediction performance among the five models. The Delong test was used to compare the differences in AUC between models pairwise. The model integrated clinical-morphological features, whole-lung radiomic features, and whole-lung image features (CMRI) performed best among the five models, and achieved the highest AUC in the internal validation set, external test set 1, and external test set 2, which were 0.886 (95% CI: 0.841-0.921), 0.830 (95%CI: 0.749-0.893) and 0.879 (95%CI: 0.712-0.968), respectively. In the above three sets, the differences in AUC between the CMRI model and other models were significant (allP< 0.05). Moreover, the accuracy of the CMRI model in the external test set 3 was 96.88%. The baseline whole-lung CT features were feasible to predict the benign and malignant of GGNs, which is helpful for more refined management ofGGNs.

Value of deep learning reconstruction of chest low-dose CT for image quality improvement and lung parenchyma assessment on lung window.

To explore the performance of low-dose computed tomography (LDCT) with deep learning reconstruction (DLR) for the improvement of image quality and assessment of lung parenchyma. Sixty patients underwent chest regular-dose CT (RDCT) followed by LDCT during the same examination. RDCT images were reconstructed with hybrid iterative reconstruction (HIR) and LDCT images were reconstructed with HIR and DLR, both using lung algorithm. Radiation exposure was recorded. Image noise, signal-to-noise ratio, and subjective image quality of normal and abnormal CT features were evaluated and compared using the Kruskal-Wallis test with Bonferroni correction. The effective radiation dose of LDCT was significantly lower than that of RDCT (0.29 ± 0.03 vs 2.05 ± 0.65 mSv, p < 0.001). The mean image noise ± standard deviation was 33.9 ± 4.7, 39.6 ± 4.3, and 31.1 ± 3.2 HU in RDCT, LDCT HIR-Strong, and LDCT DLR-Strong, respectively (p < 0.001). The overall image quality of LDCT DLR-Strong was significantly better than that of LDCT HIR-Strong (p < 0.001) and comparable to that of RDCT (p > 0.05). LDCT DLR-Strong was comparable to RDCT in evaluating solid nodules, increased attenuation, linear opacity, and airway lesions (all p > 0.05). The visualization of subsolid nodules and decreased attenuation was better with DLR than with HIR in LDCT but inferior to RDCT (all p < 0.05). LDCT DLR can effectively reduce image noise and improve image quality. LDCT DLR provides good performance for evaluating pulmonary lesions, except for subsolid nodules and decreased lung attenuation, compared to RDCT-HIR. The study prospectively evaluated the contribution of DLR applied to chest low-dose CT for image quality improvement and lung parenchyma assessment. DLR can be used to reduce radiation dose and keep image quality for several indications. • DLR enables LDCT maintaining image quality even with very low radiation doses. • Chest LDCT with DLR can be used to evaluate lung parenchymal lesions except for subsolid nodules and decreased lung attenuation. • Diagnosis of pulmonary emphysema or subsolid nodules may require higher radiation doses.

A Radiological Review of the Varied Duodenal Lesions with Histopathological Correlation

Majority of the duodenal lesions have received little attention in imaging literature in comparison with jejunum and ileum as most of them are primarily investigated with conventional endoscopy. Cross sectional imaging techniques–Multi Detector Computed Tomography (MDCT) and Magnetic Resonance Imaging (MRI) have dramatic improvement in diagnostic capabilities as they can analyse the wall of the duodenum, intraluminal content, submucosal extent of the disease, and infiltration into the surrounding viscera. Few of the duodenal lesions have overlapping imaging characteristics and it is crucial to correlate these conditions with histopathological analysis for definitive diagnosis. The main aim of this imaging review is to elucidate the diverse spectrum of duodenal lesions with emphasis on CT and MRI imaging features.

Integration and Application of Radiologic Patterns From Clinical Practice Guidelines on Idiopathic Pulmonary Fibrosis and Fibrotic Hypersensitivity Pneumonitis

Clinical practice guidelines separately describe radiologic patterns of usual interstitial pneumonia (UIP) and fibrotic hypersensitivity pneumonitis (fHP), without direction on whether or how to apply these approaches concurrently within a single patient. How can we integrate guideline-defined radiologic patterns to diagnose interstitial lung disease (ILD) and what are the pitfalls associated with described patterns that require reassessment in future guidelines? Patients from the Canadian Registry for Pulmonary Fibrosis underwent detailed reevaluation in standardized multidisciplinary discussion. CT scan features were quantified by chest radiologists masked to clinical data, and guideline-defined patterns were assigned. Clinical data then were provided to the radiologist and an ILD clinician, who jointly determined the leading diagnosis. Clinical-radiologic diagnosis in 1,593 patients was idiopathic pulmonary fibrosis (IPF) in 26%, fHP in 12%, connective tissue disease-associated ILD (CTD-ILD) in 34%, idiopathic pneumonia with autoimmune features in 12%, and unclassifiable ILD in 10%. Typical and probable UIP patterns corresponded to a diagnosis of IPF in 66%and 57%of patients, respectively. Typical fHP pattern corresponded to an fHP clinical diagnosis in 65%of patients, whereas compatible fHP was nonspecific and associated with CTD-ILD or IPAF in 48%of patients. No pattern ruled out CTD-ILD. Gas trapping affecting > 5%of lung parenchyma on expiratory imaging was an important feature broadly separating compatible and typical fHP from other patterns (sensitivity, 0.77; specificity, 0.91). An integrated approach to guideline-defined UIP and fHP patterns is feasible and supports > 5%gas trapping as an important branch point. Typical or probable UIP and typical fHP patterns have moderate predictive values for a corresponding diagnosis of IPF and fHP, although occasionally confounded by CTD-ILD; compatible fHP is nonspecific.