Low-dose liver CT: image quality and diagnostic accuracy of deep learning image reconstruction algorithm.

OBJECTIVES: To perform a comprehensive within-subject image quality analysis of abdominal CT examinations reconstructed with DLIR and to evaluate diagnostic accuracy compared to the routinely applied adaptive statistical iterative reconstruction (ASiR-V) algorithm. MATERIALS AND METHODS: Oncologic patients were prospectively enrolled and underwent contrast-enhanced CT. Images were reconstructed with DLIR with three intensity levels of reconstruction (high, medium, and low) and ASiR-V at strength levels from 10 to 100% with a 10% interval. Three radiologists characterized the lesions and two readers assessed diagnostic accuracy and calculated signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), figure of merit (FOM), and subjective image quality, the latter with a 5-point Likert scale. RESULTS: Fifty patients (mean age: 70 ± 10 years, 23 men) were enrolled and 130 liver lesions (105 benign lesions, 25 metastases) were identified. DLIR_H achieved the highest SNR and CNR, comparable to ASiR-V 100% (p ≥ .051). DLIR_M returned the highest subjective image quality (score: 5; IQR: 4-5; p ≤ .001) and significant median increase (29%) in FOM (p < .001). Differences in detection were identified only for lesions ≤ 0.5 cm: 32/33 lesions were detected with DLIR_M and 26 lesions were detected with ASiR-V 50% (p = .031). Lesion accuracy of was 93.8% (95% CI: 88.1, 97.3; 122 of 130 lesions) for DLIR and 87.7% (95% CI: 80.8, 92.8; 114 of 130 lesions) for ASiR-V 50%. CONCLUSIONS: DLIR yields superior image quality and provides higher diagnostic accuracy compared to ASiR-V in the assessment of hypovascular liver lesions, in particular for lesions ≤ 0.5 cm. CLINICAL RELEVANCE STATEMENT: Deep learning image reconstruction algorithm demonstrates higher diagnostic accuracy compared to iterative reconstruction in the identification of hypovascular liver lesions, especially for lesions ≤ 0.5 cm. KEY POINTS: • Iterative reconstruction algorithm impacts image texture, with negative effects on diagnostic capabilities. • Medium-strength deep learning image reconstruction algorithm outperforms iterative reconstruction in the diagnostic accuracy of ≤ 0.5 cm hypovascular liver lesions (93.9% vs 78.8%), also granting higher objective and subjective image quality. • Deep learning image reconstruction algorithm can be safely implemented in routine abdominal CT protocols in place of iterative reconstruction.

CT-based radiomics for prediction of therapeutic response to Everolimus in metastatic neuroendocrine tumors.

AIM: To test radiomic approach in patients with metastatic neuroendocrine tumors (NETs) treated with Everolimus, with the aim to predict progression-free survival (PFS) and death. MATERIALS AND METHODS: Twenty-five patients with metastatic neuroendocrine tumors, 15/25 pancreatic (60%), 9/25 ileal (36%), 1/25 lung (4%), were retrospectively enrolled between August 2013 and December 2020. All patients underwent contrast-enhanced CT before starting Everolimus, histological diagnosis, tumor grading, PFS, overall survival (OS), death, and clinical data collected. Population was divided into two groups: responders (PFS ≤ 11 months) and non-responders (PFS > 11 months). 3D segmentation was performed on whole liver of naïve CT scans in arterial and venous phases, using a dedicated software (3DSlicer v4.10.2). A total of 107 radiomic features were extracted and compared between two groups (T test or Mann-Whitney), radiomics performance assessed with receiver operating characteristic curve, Kaplan-Meyer curves used for survival analysis, univariate and multivariate logistic regression performed to predict death, and interobserver variability assessed. All significant radiomic comparisons were validated by using a synthetic external cohort. P < 0.05 is considered significant. RESULTS: 15/25 patients were classified as responders (median PFS 25 months and OS 29 months) and 10/25 as non-responders (median PFS 4.5 months and OS 23 months). Among radiomic parameters, Correlation and Imc1 showed significant differences between two groups (P < 0.05) with the best performance (internal cohort AUC 0.86-0.84, P < 0.0001; external cohort AUC 0.84-0.90; P < 0.0001). Correlation < 0.21 resulted correlated with death at Kaplan-Meyer analysis (P = 0.02). Univariate analysis showed three radiomic features independently correlated with death, and in multivariate analysis radiomic model showed good performance with AUC 0.87, sensitivity 100%, and specificity 66.7%. Three features achieved 0.77 ≤ ICC < 0.83 and one ICC = 0.92. CONCLUSIONS: In patients affected by metastatic NETs eligible for Everolimus treatment, radiomics could be used as imaging biomarker able to predict PFS and death.

Artificial intelligence based image quality enhancement in liver MRI: a quantitative and qualitative evaluation.

PURPOSE: To compare liver MRI with AIR Recon Deep Learning™(ARDL) algorithm applied and turned-off (NON-DL) with conventional high-resolution acquisition (NAÏVE) sequences, in terms of quantitative and qualitative image analysis and scanning time. MATERIAL AND METHODS: This prospective study included fifty consecutive volunteers (31 female, mean age 55.5 ± 20 years) from September to November 2021. 1.5 T MRI was performed and included three sets of images: axial single-shot fast spin-echo (SSFSE) T2 images, diffusion-weighted images(DWI) and apparent diffusion coefficient(ADC) maps acquired with both ARDL and NAÏVE protocol; the NON-DL images, were also assessed. Two radiologists in consensus drew fixed regions of interest in liver parenchyma to calculate signal-to-noise-ratio (SNR) and contrast to-noise-ratio (CNR). Subjective image quality was assessed by two other radiologists independently with a five-point Likert scale. Acquisition time was recorded. RESULTS: SSFSE T2 objective analysis showed higher SNR and CNR for ARDL vs NAÏVE, ARDL vs NON-DL(all P < 0.013). Regarding DWI, no differences were found for SNR with ARDL vs NAÏVE and, ARDL vs NON-DL (all P > 0.2517).CNR was higher for ARDL vs NON-DL(P = 0.0170), whereas no differences were found between ARDL and NAÏVE(P = 1). No differences were observed for all three comparisons, in terms of SNR and CNR, for ADC maps (all P > 0.32). Qualitative analysis for all sequences showed better overall image quality for ARDL with lower truncation artifacts, higher sharpness and contrast (all P < 0.0070) with excellent inter-rater agreement (k ≥ 0.8143). Acquisition time was lower in ARDL sequences compared to NAÏVE (SSFSE T2 = 19.08 ± 2.5 s vs. 24.1 ± 2 s and DWI = 207.3 ± 54 s vs. 513.6 ± 98.6 s, all P < 0.0001). CONCLUSION: ARDL applied on upper abdomen showed overall better image quality and reduced scanning time compared with NAÏVE protocol.

3D pelvimetry and biometric measurements: a surgical perspective for colorectal resections.

PURPOSE: Male sex, high BMI, narrow pelvis, and bulky mesorectum were acknowledged as clinical variables correlated with a difficult pelvic dissection in colorectal surgery. This paper aimed at comparing pelvic biometric measurements in female and male patients and at providing a perspective on how pelvimetry segmentation may help in visualizing mesorectal distribution. METHODS: A 3D software was used for segmentation of DICOM data of consecutive patients aged 60 years, who underwent elective abdominal CT scan. The following measurements were estimated: pelvic inlet, outlet, and depth; pubic tubercle height; distances from the promontory to the coccyx and to S3/S4; distance from S3/S4 to coccyx's tip; ischial spines distance; pelvic tilt; offset angle; pelvic inlet angle; angle between the inlet/sacral promontory/coccyx; angle between the promontory/coccyx/pelvic outlet; S3 angle; and pelvic inlet to pelvic depth ratio. The measurements were compared in males and females using statistical analyses. RESULTS: Two-hundred patients (M/F 1:1) were analyzed. Out of 21 pelvimetry measurements, 19 of them documented a significant mean difference between groups. Specifically, female patients had a significantly wider pelvic inlet and outlet but a shorter pelvic depth, and promontory/sacral/coccyx distances, resulting in an augmented inlet/depth ratio when comparing with males (p < 0.0001). The sole exceptions were the straight conjugate (p = 0.06) and S3 angle (p = 0.17). 3D segmentation provided a perspective of the mesorectum distribution according to the pelvic shape. CONCLUSION: Significant differences in the structure of pelvis exist in males and females. Surgeons must be aware of the pelvic shape when approaching the rectum.

Results of First-Round of Surveillance in Individuals at High-Risk of Pancreatic Cancer from the AISP (Italian Association for the Study of the Pancreas) Registry.

INTRODUCTION: Surveillance programs on high-risk individuals (HRIs) can detect pre-malignant lesions or early pancreatic cancer (PC). We report the results of the first screening round of the Italian multicenter program supported by the Italian Association for the study of the Pancreas (AISP). METHODS: The multicenter surveillance program included asymptomatic HRIs with familial (FPC) or genetic frailty (GS: BRCA1/2, p16/CDKN2A, STK11/LKB1or PRSS1, mutated genes) predisposition to PC. The surveillance program included at least an annual magnetic resonance cholangio pancreatography (MRCP). Endoscopic ultrasound (EUS) was proposed to patients who refused or could not be submitted to MRCP. RESULTS: One-hundreds eighty-seven HRIs underwent a first-round screening examination with MRCP (174; 93.1%) or EUS (13; 6.9%) from September 2015 to March 2018.The mean age was 51 years (range 21-80).One-hundreds sixty-five (88.2%) FPC and 22 (11.8%) GF HRIs were included. MRCP detected 28 (14.9%) presumed branch-duct intraductal papillary mucinous neoplasms (IPMN), 1 invasive carcinoma/IPMN and one low-grade mixed-type IPMN, respectively. EUS detected 4 PC (2.1%): 1 was resected, 1 was found locally advanced intraoperatively, and 2 were metastatic. Age > 50 (OR 3.3, 95%CI 1.4-8), smoking habit (OR 2.8, 95%CI 1.1-7.5), and having > 2 relatives with PC (OR 2.7, 95%CI 1.1-6.4) were independently associated with detection of pre-malignant and malignant lesions. The diagnostic yield for MRCP/EUS was 24% for cystic lesions. The overall rate of surgery was 2.6% with nil mortality. DISCUSSION: The rate of malignancies found in this cohort was high (2.6%). According to the International Cancer of the Pancreas Screening Consortium the screening goal achievement was high (1%).

Prognostic impact of tumour burden in stage IV neuroendocrine neoplasia: A comparison between pancreatic and gastrointestinal localizations.

BACKGROUND: Although prognosis of NENs is affected by several features including tumour burden, the specific role of this factor in pancreatic NENs (PanNENs) and gastrointestinal NENs (GI NENs) is not well established. AIM: To compare the prognostic role of tumour burden in PanNENs and GI NENs. PATIENTS AND METHODS: This study was a retrospective analysis of stage IV PanNENs and GI NENs. Tumours were classified based on liver tumour volume (<25% or >25%). Overall survival as assessed by Kaplan-Meier curves, and Cox proportional hazards method was used to perform risk factor analysis. RESULTS: The analysis included 300 patients, including 166 panNENs (55.3%) and 134 GI NENs (44.7%). A total of 158 patients (52.7%) had G2 tumours, 107 had G1 tumours (35.7%), and 35 had G3 tumours (11.6%). Tumour liver involvement >25% was observed in 187 patients (62.3%): 106 PanNENs (56.7%), and 81 GI NENs (43.3%) (p = 0.551). Bone metastases were present in 45 patients (15%): 22 PanNENs (13.2%) and 23 GI NENs (17.1%) (p = 0.416). Characteristics of the PanNENs, including: grading (G2 vs G1, HR = 3.7; G3 vs G1, HR = 16.40), liver involvement > 25% (HR = 3.09), and bone metastases (HR = 2.27) were independent predictors for poor survival, whereas the only significant risk factor in GI NENs was grading (G2 vs G1, HR = 4.36; G3 vs G1, HR = 8.60). CONCLUSIONS: PanNENs and GI NENs have different risk profiles. Liver tumour volume and the presence of bone metastases significantly affect survival in patients with PanNENs but has no impact on the clinical outcomes of GI NENs.

Clinical Usefulness of 18F-Fluorodeoxyglucose Positron Emission Tomography in the Diagnostic Algorithm of Advanced Entero-Pancreatic Neuroendocrine Neoplasms.

BACKGROUND: The role of 18F-fluorodeoxyglucose positron emission tomography (18F-FDG PET) in the diagnostic algorithm of entero-pancreatic neuroendocrine neoplasms (EP NENs) is unclear because most available data derive from heterogeneous populations in terms of tumor biology and disease status at time of examination. The aim of this study was to determine the ability of 18F-FDG PET to identify patients with more aggressive disease among those with advanced EP NENs. Subjects, Materials, and Methods . Patients with advanced EP NENs and known disease status (progressive disease [PD] or stable disease [SD]) according to imaging procedures, who received 18F-FDG PET and computed tomography scans during a time frame of 1 month, were included. RESULTS: A total of 93 patients, including 69 patients with pancreatic NENs and 24 patients with small-intestine NENs, were included. At the time of study entry, 64 patients (68.8%) had PD, and the remaining 29 patients (31.2%) had SD. A total of 62 patients (66.7%) had positive 18F-FDG PET, whereas 18F-FDG PET was negative in the remaining 31 patients (33.3%). Overall, 18F-FDG PET sensitivity and specificity to detect PD were 90.6% and 86.2%, respectively, resulting in a diagnostic accuracy of 89.2%. A positive 18F-FDG PET was significantly associated with PD at the time of study entry (p < .0001 at multivariate analysis). Although a higher proportion of 18F-FDG PET-positive examinations were observed in patients with higher tumor grade (p = .01), 53.8% of patients with grade 1 neuroendocrine tumors (NETs) had positive 18F-FDG PET, and 37.5% of patients with grade 2 NETs had negative 18F-FDG PET. Overall survival was significantly shorter in 18F-FDG PET-positive patients (median: 60 months) in comparison with 18F-FDG PET-negative patients (median not reached; p = .008). CONCLUSION: 18F-FDG PET has a high diagnostic accuracy to identify progression of disease with unfavorable clinical outcome in patients with advanced EP NENs. Knowledge of disease status and G grading are key factors for physicians to better select patients for whom 18F-FDG PET is clinically useful. IMPLICATIONS FOR PRACTICE: The findings of the present study may help physicians dealing with advanced neuroendocrine neoplasms to select patients for whom 18F-fluorodeoxyglucose positron emission tomography is useful to predict poor clinical outcome.

Evaluation of the Relationships Between Computed Tomography Features, Pathological Findings, and Prognostic Risk Assessment in Gastrointestinal Stromal Tumors.

OBJECTIVES: The aim of this study was to correlate computed tomography (CT) findings with pathology in gastrointestinal stromal tumors (GISTs). METHODS: A retrospective evaluation of CT images of 44 patients with GISTs was performed. Computed tomography findings analyzed were location, size, margins, degree and pattern of contrast enhancement, angiogenesis, necrosis, signs of invasion, peritoneal effusion, peritoneal implants, surface ulceration, and calcifications.Associations between CT features and mitotic rate, Miettinen classes of risk, lesions size, and among CT features were investigated. χ Test and Fisher test were performed. RESULTS: Mitotic rate was associated with margins (P = 0.016) and with adjacent organ invasion (P = 0.043). Pattern of contrast enhancement (P = 0.002), angiogenesis (P = 0.006), necrosis (P = 0.006), invasion of adjacent organs (P = 0.011), and margins (P = 0.006) were associated with classes of risk. Several associations (P < 0.05) between lesion size and CT features and among all the investigated CT features were found. CONCLUSIONS: Computed tomography features could reflect GIST biology being associated with the mitotic rate and with classes of risk.

Sarcopenia predicts reduced survival in patients with hepatocellular carcinoma at first diagnosis.

 Background. Sarcopenia is a complication and independent risk factor for mortality in patients with liver cirrhosis. AIM: To assess the prevalence and influence of sarcopenia on overall survival in a cohort of cirrhotic patients with hepatocellular carcinoma managed in a tertiary center. MATERIAL AND METHODS: Abdominal computed tomography of 92 consecutive hepatocellular carcinoma cirrhotic patients, enrolled and followed from 2004 to 2014, were retrospectively studied with a software analyzing the cross-sectional areas of muscles at third lumbar vertebra level. Data was normalized for height, skeletal muscle index (SMI) calculated and presence of Sarcopenia measured. Sarcopenia was defined by SMI ≤ 41 cm2/m2 for women and ≤ 53 cm2/m2 for men with body mass index (BMI) ≥ 25, and ≤ 43 cm2/m2 for men and women with BMI < 25, respectively. RESULTS: Median age at diagnosis was 71.9 years (30.7-86.4) and BMI 24.7 (17.5-36.7), comparable in women 23.1, (17.5-36.7) and men 24.7 (18.4-36.7). A class of CHILD score and BCLC A prevailed (55.4% and 41.3%, respectively); metastatic disease was found in 12% of cases. Sarcopenia was present in 40.2% of cases, mostly in females (62.9%; p = 0.005). Mean overall survival was reduced in sarcopenic patients, 66 (95% CI 47 to 84) vs. 123 (95% CI 98 to 150) weeks (p = 0.001). At multivariate analysis, sarcopenia was a predictor of reduced overall survival, independent of age (p = 0.0027). CONCLUSIONS: This retrospective study shows high prevalence of sarcopenia among cirrhotic patients with hepatocellular carcinoma. Presence of sarcopenia was identified as independent predictor of reduced overall survival. As easily measurable by CT, sarcopenia should be determined for prognostic purposes in this patient population.

The role of contrast enhanced computed tomography in the diagnosis of necrotizing fasciitis and comparison with the laboratory risk indicator for necrotizing fasciitis (LRINEC).

PURPOSE: To evaluate the diagnostic efficacy of contrast enhanced computed tomography (CECT) in emergency departments for diagnosis of necrotizing fasciitis (NF) and for differential diagnosis of other musculoskeletal infections; to correlate radiological findings with the laboratory risk indicator for necrotizing fasciitis (LRINEC). MATERIALS AND METHODS: 7 radiological parameters to be analysed on CECT scans were established, exams of 36 patients with proven diagnosis of NF (n 12) and other musculoskeletal infections (n 24) were retrospectively reviewed; LRINEC score was calculated. Fisher's test and Spearman's and Kendall's coefficients of rank correlations were performed. RESULTS: Two parameters were found to be strongly associated with the diagnosis of NF: involvement of the fascia (Spearman's ρ of 0.888, p < 0.001) and lack of fascial enhancement (Spearman's ρ of 0.672, p < 0.001). LRINEC score did not show strong association with the presence of fasciitis NF (Spearman's ρ of 0.490, p = 0.0024). CONCLUSION: Computed tomography (CT) parameters, which are significantly associated with the diagnosis of NF, are the involvement of the fascia and its lack of enhancement; LRINEC score could be high (>5) also in other musculoskeletal infections. Final diagnosis of necrosis among the fascia is surgical. Presence of gas is not a specific sign of necrotizing fasciitis being present in other musculoskeletal infections. CT could easily discriminate NF from other musculoskeletal infections, adds an important value to clinical and laboratory tests in diagnosis of NF in an emergency context when magnetic resonance imaging, which is superior to CT in this discernment, could not be performed.

Gadoxetic acid-enhanced MRI for hepatocellular carcinoma and hypointense nodule observed in the hepatobiliary phase.

PURPOSE: The aim of our study was to evaluate the diagnostic accuracy of gadoxetic acid-enhanced magnetic resonance (MR) imaging both in the detection of hepatocellular carcinoma (HCC) and precancerous lesions and in the assessment of their evolution. MATERIALS AND METHODS: A retrospective study was undertaken on 56 patients with chronic liver disease and suspected liver lesions. We evaluated the number, size and signal intensity of the nodules on dynamic and hepatobiliary MR images. Follow-up studies were carried out every 3 months. Statistical analysis was performed using the Fisher's exact test. RESULTS: A total of 120 nodules were identified in 41 patients. Of these, 92/120 nodules (76.6%; mean diameter 18.4 mm) showed the typical HCC vascular pattern: 90/92 nodules appeared hypointense and 2/92 were hyperintense on hepatobiliary phase images. An additional 28/120 hypointense, nonhypervascular nodules (23.3%; mean diameter 11 mm) were detected on hepatobiliary phase images, 15 of which showed hypointensity also on the equilibrium phase images. During the 3- to 12-month follow-up, 14/28 nodules (mean diameter 13.3 mm) developed the typical vascular pattern of HCC. CONCLUSIONS: Gadoxetic acid-enhanced MR imaging is useful for detecting HCC as well as hypovascular nodules with potential progression to HCC. Lesions measuring more than 10 mm in diameter are at higher risk of developing into HCC (p = 0.0128).

Gastrointestinal bleeding in children: diagnostic approach.

Different conditions may underlie gastrointestinal bleeding (GIB) in children. The estimated prevalence of GIB in children is 6.4%, with spontaneous resolution in approximately 80% of cases. Nonetheless, the initial approach plays a pivotal role in determining the prognosis. The priority is the stabilization of hemodynamic status, followed by a systematic diagnostic approach. GIB can originate from either upper or lower gastrointestinal tract, leading to a broad differential diagnosis in infants and children. This includes benign and self-limiting disorders, alongside serious conditions necessitating immediate treatment. We performed a nonsystematic review of the literature, in order to describe the variety of conditions responsible for GIB in pediatric patients and to outline diagnostic pathways according to patients' age, suspected site of bleeding and type of bleeding which can help pediatricians in clinical practice. Diagnostic modalities may include esophagogastroduodenoscopy and colonoscopy, abdominal ultrasonography or computed tomography and, when necessary, magnetic resonance imaging. In this review, we critically assess these procedures, emphasizing their respective advantages and limitations concerning specific clinical scenarios.

Liver Magnetic Resonance Elastography: Focus on Methodology, Technique, and Feasibility.

Magnetic resonance elastography (MRE) is an imaging technique that combines low-frequency mechanical vibrations with magnetic resonance imaging to create visual maps and quantify liver parenchyma stiffness. As in recent years, diffuse liver diseases have become highly prevalent worldwide and could lead to a chronic condition with different stages of fibrosis. There is a strong necessity for a non-invasive, highly accurate, and standardised quantitative assessment to evaluate and manage patients with different stages of fibrosis from diagnosis to follow-up, as the actual reference standard for the diagnosis and staging of liver fibrosis is biopsy, an invasive method with possible peri-procedural complications and sampling errors. MRE could quantitatively evaluate liver stiffness, as it is a rapid and repeatable method with high specificity and sensitivity. MRE is based on the propagation of mechanical shear waves through the liver tissue that are directly proportional to the organ's stiffness, expressed in kilopascals (kPa). To obtain a valid assessment of the real hepatic stiffness values, it is mandatory to obtain a high-quality examination. To understand the pearls and pitfalls of MRE, in this review, we describe our experience after one year of performing MRE from indications and patient preparation to acquisition, quality control, and image analysis.

Is CT Radiomics Superior to Morphological Evaluation for pN0 Characterization? A Pilot Study in Colon Cancer.

The aim of this study was to compare CT radiomics and morphological features when assessing benign lymph nodes (LNs) in colon cancer (CC). This retrospective study included 100 CC patients (test cohort) who underwent a preoperative CT examination and were diagnosed as pN0 after surgery. Regional LNs were scored with a morphological Likert scale (NODE-SCORE) and divided into two groups: low likelihood (LLM: 0-2 points) and high likelihood (HLM: 3-7 points) of malignancy. The T-test and the Mann-Whitney test were used to compare 107 radiomic features extracted from the two groups. Radiomic features were also extracted from primary lesions (PLs), and the receiver operating characteristic (ROC) was used to test a LN/PL ratio when assessing the LN's status identified with radiomics and with the NODE-SCORE. An amount of 337 LNs were divided into 167 with LLM and 170 with HLM. Radiomics showed 15/107 features, with a significant difference (p < 0.02) between the two groups. The comparison of selected features between 81 PLs and the corresponding LNs showed all significant differences (p < 0.0001). According to the LN/PL ratio, the selected features recognized a higher number of LNs than the NODE-SCORE (p < 0.001). On validation of the cohort of 20 patients (10 pN0, 10 pN2), significant ROC curves were obtained for LN/PL busyness (AUC = 0.91; 0.69-0.99; 95% C.I.; and p < 0.001) and for LN/PL dependence entropy (AUC = 0.76; 0.52-0.92; 95% C.I.; and p = 0.03). The radiomics ratio between CC and LNs is more accurate for noninvasively discriminating benign LNs compared to CT morphological features.

Development and Validation of Artificial-Intelligence-Based Radiomics Model Using Computed Tomography Features for Preoperative Risk Stratification of Gastrointestinal Stromal Tumors.

BACKGROUND: preoperative risk assessment of gastrointestinal stromal tumors (GISTS) is required for optimal and personalized treatment planning. Radiomics features are promising tools to predict risk assessment. The purpose of this study is to develop and validate an artificial intelligence classification algorithm, based on CT features, to define GIST's prognosis as determined by the Miettinen classification. METHODS: patients with histological diagnosis of GIST and CT studies were retrospectively enrolled. Eight morphologic and 30 texture CT features were extracted from each tumor and combined to obtain three models (morphologic, texture and combined). Data were analyzed using a machine learning classification (WEKA). For each classification process, sensitivity, specificity, accuracy and area under the curve were evaluated. Inter- and intra-reader agreement were also calculated. RESULTS: 52 patients were evaluated. In the validation population, highest performances were obtained by the combined model (SE 85.7%, SP 90.9%, ACC 88.8%, and AUC 0.954) followed by the morphologic (SE 66.6%, SP 81.8%, ACC 76.4%, and AUC 0.742) and texture (SE 50%, SP 72.7%, ACC 64.7%, and AUC 0.613) models. Reproducibility was high of all manual evaluations. CONCLUSIONS: the AI-based radiomics model using a CT feature demonstrates good predictive performance for preoperative risk stratification of GISTs.

Performance of Machine Learning and Texture Analysis for Predicting Response to Neoadjuvant Chemoradiotherapy in Locally Advanced Rectal Cancer with 3T MRI.

Background: To evaluate the diagnostic performance of a Machine Learning (ML) algorithm based on Texture Analysis (TA) parameters in the prediction of Pathological Complete Response (pCR) to Neoadjuvant Chemoradiotherapy (nChRT) in Locally Advanced Rectal Cancer (LARC) patients. Methods: LARC patients were prospectively enrolled to undergo pre- and post-nChRT 3T MRI for initial loco-regional staging. TA was performed on axial T2-Weighted Images (T2-WI) to extract specific parameters, including skewness, kurtosis, entropy, and mean of positive pixels. For the assessment of TA parameter diagnostic performance, all patients underwent complete surgical resection, which served as a reference standard. ROC curve analysis was carried out to determine the discriminatory accuracy of each quantitative TA parameter to predict pCR. A ML-based decisional tree was implemented combining all TA parameters in order to improve diagnostic accuracy. Results: Forty patients were considered for final study population. Entropy, kurtosis and MPP showed statistically significant differences before and after nChRT in patients with pCR; in particular, when patients with Pathological Partial Response (pPR) and/or Pathological Non-Response (pNR) were considered, entropy and skewness showed significant differences before and after nChRT (all p < 0.05). In terms of absolute value changes, pre- and post-nChRT entropy, and kurtosis showed significant differences (0.31 ± 0.35, in pCR, −0.02 ± 1.28 in pPR/pNR, (p = 0.04); 1.87 ± 2.19, in pCR, −0.06 ± 3.78 in pPR/pNR (p = 0.0005); 107.91 ± 274.40, in pCR, −28.33 ± 202.91 in pPR/pNR, (p = 0.004), respectively). According to ROC curve analysis, pre-treatment kurtosis with an optimal cut-off value of ≤3.29 was defined as the best discriminative parameter, resulting in a sensitivity and specificity in predicting pCR of 81.5% and 61.5%, respectively. Conclusions: TA parameters extracted from T2-WI MRI images could play a key role as imaging biomarkers in the prediction of response to nChRT in LARC patients. ML algorithms can be used to efficiently combine all TA parameters in order to improve diagnostic accuracy.

Rectal cancer response to neoadjuvant chemoradiotherapy evaluated with MRI: Development and validation of a classification algorithm.

OBJECTIVE: The aim of this study was to develop and validate a decision support model using data mining algorithms, based on morphologic features derived from MRI images, to discriminate between complete responders (CR) and non-complete responders (NCR) patients after neoadjuvant chemoradiotherapy (CRT), in a population of patients with locally advanced rectal cancer (LARC). METHODS: Two populations were retrospectively enrolled: group A (65 patients) was used to train a data mining decision tree algorithm whereas group B (30 patients) was used to validate it. All patients underwent surgery; according to the histology evaluation, patients were divided in CR and NCR. Staging and restaging MRI examinations were retrospectively analysed and seven parameters were considered for data mining classification. Five different classification methods were tested and evaluated in terms of sensitivity, specificity, accuracy and AUC in order to identify the classification model able to achieve the best performance. The best classification algorithm was subsequently applied to group B for validation: sensitivity, specificity, positive and negative predictive value, accuracy and ROC curve were calculated. Inter and intra-reader agreement were calculated. RESULTS: Four features were selected for the development of the classification algorithm: MRI tumor regression grade (MR-TRG), staging volume (SV), tumor volume reduction rate (TVRR) and signal intensity reduction rate (SIRR). The decision tree J48 showed the highest efficiency: when applied to group B, all the CR and 18/21 NCR were correctly classified (sensitivity 85.71%, specificity 100%, PPV 100%, NPV 94.2%, accuracy 95.7%, AUC 0.833). Both inter- and intra-reader evaluation showed good agreement (κ > 0.6). CONCLUSIONS: The proposed decision support model may help in distinguishing between CR and NCR patients with LARC after CRT.

Radiomic Cancer Hallmarks to Identify High-Risk Patients in Non-Metastatic Colon Cancer.

The study was aimed to develop a radiomic model able to identify high-risk colon cancer by analyzing pre-operative CT scans. The study population comprised 148 patients: 108 with non-metastatic colon cancer were retrospectively enrolled from January 2015 to June 2020, and 40 patients were used as the external validation cohort. The population was divided into two groups­High-risk and No-risk­following the presence of at least one high-risk clinical factor. All patients had baseline CT scans, and 3D cancer segmentation was performed on the portal phase by two expert radiologists using open-source software (3DSlicer v4.10.2). Among the 107 radiomic features extracted, stable features were selected to evaluate the inter-class correlation (ICC) (cut-off ICC > 0.8). Stable features were compared between the two groups (T-test or Mann−Whitney), and the significant features were selected for univariate and multivariate logistic regression to build a predictive radiomic model. The radiomic model was then validated with an external cohort. In total, 58/108 were classified as High-risk and 50/108 as No-risk. A total of 35 radiomic features were stable (0.81 ≤ ICC < 0.92). Among these, 28 features were significantly different between the two groups (p < 0.05), and only 9 features were selected to build the radiomic model. The radiomic model yielded an AUC of 0.73 in the internal cohort and 0.75 in the external cohort. In conclusion, the radiomic model could be seen as a performant, non-invasive imaging tool to properly stratify colon cancers with high-risk disease.

Gastrointestinal stromal tumors: correlation between symptoms at presentation, tumor location and prognostic factors in 47 consecutive patients.

BACKGROUND: Gastrointestinal stromal tumors (GIST) are mesenchymal tumors of the gastrointestinal tract, usually kit-positive, that are believed to originate from interstitial cell of Cajal, or their related stem cells. The most common clinical presentation of these tumors is gastrointestinal bleeding, otherwise they may cause intestinal obstruction, abdominal pain, a palpable mass, or can be incidentally detected during surgery or endoscopic/radiological procedures. Prognosis is related to the size of the tumor and to the mitotic rate; other prognostic factors are tumor location, tumor resection margins, tumor rupture, and c-kit mutation that may interfere with molecular target therapy efficacy. AIM: Primary aim of this study was to report our experience regarding GIST patients, correlating symptoms at presentation with tumor localization and risk factors. PATIENTS AND METHODS: 47 consecutive patients undergone to surgical resection for GISTs were enrolled in a prospective study from December 1999 to March 2009. Patient's clinical and pathological features were collected and analysed. RESULTS: The most common symptom was abdominal pain. Bleeding in the digestive tract and abdominal pain were more frequent in gastric GISTs (58% and 61%); acute abdominal symptoms were more frequent in jejunal and ileal GISTs (40% and 60%), p < 0.05. We reported a mild correlation between the mitotic rate index and symptoms at presentation (p 0.074): this correlation was stronger if GISTs causing "acute abdominal symptoms" were compared with GISTs causing "abdominal pain" as main symptom (p 0.039) and with "incidental" GISTs (p 0.022).We observed an higher prevalence of symptomatic patients in the "high risk/malignant group" of both the Fletcher's and Miettines's classification (p < 0.05). CONCLUSION: According with our findings symptoms correlate to tumor location, to class risk criteria as mitotic index and risk classifications, however we cannot conclude that symptoms are per se predictive of survival or patient's outcome.

Efficacy of Lutetium-Peptide Receptor Radionuclide Therapy in Inducing Prolonged Tumour Regression in Small-Bowel Neuroendocrine Tumours: A Case of Favourable Response to Retreatment after Initial Objective Response.

INTRODUCTION: The efficacy of 177Lu-Dotatate was shown in the NETTER-1 trial, an international, open-label, multicentre phase III clinical trial that evaluated the safety and efficacy of 177Lu-Dotatate in patients with well-differentiated, advanced midgut neuroendocrine tumours (NETs) with evidence of disease progression. Recently, retreatment with peptide receptor radionuclide therapy (PRRT) has been proposed as a valid therapeutic option in patients without other effective options who had responded to initial PRRT; however, data on this therapeutic option are still inadequate. CASE REPORT: In this report, we present the case of a patient who achieved a delayed complete radiological response after initial 177Lu-Dotatate treatment and who had a complete tumour response with PRRT retreatment 5 years later. CONCLUSIONS: This case report shows that, although rare, a complete, prolonged tumour response may occur in patients with advanced small-bowel NETs receiving PRRT. Retreatment with PRRT may be a valid option in cases of subsequent disease recurrence.