Application of multi-spectral CT imaging in Crohn's disease: a systematic review.

BACKGROUND: No quantitative computed tomography (CT) biomarker is actually sufficiently accurate to assess Crohn's disease (CD) lesion activity, with adequate precision to guide clinical decisions. PURPOSE: To assess the available literature on the use of iodine concentration (IC), from multi-spectral CT acquisition, as a quantitative parameter able to distinguish healthy from affected bowel and assess CD bowel activity and heterogeneity of activity along the involved segments. MATERIAL AND METHODS: A literature search was conducted to identify original research studies published up to February 2022. The inclusion criteria were original research papers (>10 human participants), English language publications, focus on dual-energy CT (DECT) of CD with iodine quantification (IQ) as an outcome measure. The exclusion criteria were animal-only studies, languages other than English, review articles, case reports, correspondence, and study populations <10 patients. RESULTS: Nine studies were included in this review; all of which showed a strong correlation between IC measurements and CD activity markers, such as CD activity index (CDAI), endoscopy findings and simple endoscopic score for Crohn's disease (SES-CD), and routine CT enterography (CTE) signs and histopathologic score. Statistically significant differences in IC were reported between affected bowel segments and healthy ones (higher P value was P < 0.001), normal segments and those with active inflammation (P < 0.0001) as well as between patients with active disease and those in remission (P < 0.001). CONCLUSION: The mean normalized IC at DECTE could be a reliable tool in assisting radiologists in the diagnosis, classification and grading of CD activity.

Prognostic findings for ICU admission in patients with COVID-19 pneumonia: baseline and follow-up chest CT and the added value of artificial intelligence.

Infection with SARS-CoV-2 has dominated discussion and caused global healthcare and economic crisis over the past 18 months. Coronavirus disease 19 (COVID-19) causes mild-to-moderate symptoms in most individuals. However, rapid deterioration to severe disease with or without acute respiratory distress syndrome (ARDS) can occur within 1-2 weeks from the onset of symptoms in a proportion of patients. Early identification by risk stratifying such patients who are at risk of severe complications of COVID-19 is of great clinical importance. Computed tomography (CT) is widely available and offers the potential for fast triage, robust, rapid, and minimally invasive diagnosis: Ground glass opacities (GGO), crazy-paving pattern (GGO with superimposed septal thickening), and consolidation are the most common chest CT findings in COVID pneumonia. There is growing interest in the prognostic value of baseline chest CT since an early risk stratification of patients with COVID-19 would allow for better resource allocation and could help improve outcomes. Recent studies have demonstrated the utility of baseline chest CT to predict intensive care unit (ICU) admission in patients with COVID-19. Furthermore, developments and progress integrating artificial intelligence (AI) with computer-aided design (CAD) software for diagnostic imaging allow for objective, unbiased, and rapid assessment of CT images.

Treatment-related Focal Nodular Hyperplasia Mimicking Liver Metastases.

A 7-year-old patient with a stage 4 neuroblastoma underwent chemotherapy, surgery, myeloablative therapy, external beam radiotherapy, and Isoretinoin treatment. A posttreatment magnetic resonance examination performed administering gadoteric acid as contrast agent showed 2 new hypervascular hepatic lesions, suspicious for metastases. A second magnetic resonance imaging performed using a liver-specific contrast medium (gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid) demonstrated that these lesions were consistent with treatment-related focal nodular hyperplasia.

Quantification of Iodine Concentration Using Single-Source Dual-Energy Computed Tomography in a Calf Liver.

OBJECTIVE: To evaluate the accuracy of single-source dual-energy computed tomography (ssDECT) in iodine quantification using various segmentation methods in an ex vivo model. METHODS: Ten sausages, injected with variable quantities of iodinated contrast, were inserted into 2 livers and scanned with ssDECT. Material density iodine images were reconstructed. Three radiologists segmented each sausage. Iodine concentration, volume, and absolute quantity were measured. Agreement between the measured and injected iodine was assessed with the concordance correlation coefficient (CCC). Intrareader agreement was assessed using the intraclass correlation coefficient (ICC). RESULTS: Air bubbles were observed in sausage (IX). Sausage (X) was within the same view as hyper-attenuating markers used for localization. With IX and X excluded, CCC and ICC were greater than 0.98 and greater than 0.88. When included, CCC and ICC were greater than 0.94 and greater than 0.79. CONCLUSIONS: Iodine quantification was reproducible and precise. However, accuracy reduced in sausages consisting of air filled cavities and within the same view as hyperattenuating markers.

Neuro-radiological features can predict hypopituitarism in primary autoimmune hypophysitis.

Primary autoimmune hypophysitis (PAH) is considered an underdiagnosed disease, due to the difficulty in reaching a definitive diagnosis. PAH clinical diagnosis remains an exclusion diagnosis. We aimed to correlate PAH neuro-radiological signs to endocrine features and evaluate their prognostic role. 24 PAH cases were enrolled and classified according to neuro-radiological signs: in 12 adeno-hypophysitis (AHs), 8 infundibulo-neuro-hypophysitis (INHs) and 4 pan-hypophysitis (PHs). Secondary hypogonadism developed more frequently in INHs as compared to AHs (54.5% vs. 27.3%, p = 0.05), without no difference with PHs (p = 0.6). Diabetes insipidus occurred more frequently in INHs cases (72.7%, p < 0.001) and in PHs cases (27.3%, p = 0.007), as compared to AHs cases (0%). Similarly, all cases of GHD occurred in INHs (100%) as compared to AHs (0%, p < 0.001) and PHs (0%, p < 0.001). The pituitary stalk (PS) showed a pseudo-triangular shape (larger at the optical chiasma) in INHs and a pseudo-cylindrical shape (larger both at the optical chiasma and at the pituitary insertion) in PHs. The PS pseudo-triangular shape correlated to the occurrence of GHD and diabetes insipidus (p < 0.001/p = 0.03). At the 1-year follow-up, improvement of baseline radiological features positively correlated with the loss of the neuro-pituitary "bright spot" on T1-weighted images (OR 0.16; 95% CI 0.03-0.9 p = 0.02) and with a PS diameter at the optical chiasma level larger than 4.1 mm (AUC 0.97, sensibility 80%, specificity 100%, OR 6; 95% CI1.1-28.8, p = 0.01) Our data suggest that neuro-radiological PAH classification in PH, AH and INH can predict pituitary dysfunction and that some neuro-radiological features, such as the pituitary stalk diameter and the loss of the neuro-pituitary bright spot on T1w images can play a role as a positive prognostic marker of the radiological hypophysitis outcome.

An Overview of Diagnosis of Primary Autoimmune Hypophysitis in a Prospective Single-Center Experience.

BACKGROUND: Autoimmune hypophysitis is a rare disease with a natural progression that is not well known. AIM: To collect representative data on clinical features of autoimmune hypophysitis and better characterize the disease. PATIENTS AND METHODS: A prospective single-center study was designed. Autoimmune hypophysitis-affected patients evaluated from 2011 at our tertiary care Pituitary Unit were enrolled. After ruling out other pituitary masses and secondary causes of hypophysitis, autoimmune hypophysitis was the diagnosis of exclusion. Autoimmune hypophysitis was classified as adenohypophysitis, panhypophysitis, and infundibuloneurohypophysitis according to clinical and neuroradiological findings. RESULTS: A total of 21 patients met the inclusion criteria: 9 were diagnosed with adenohypophysitis, 4 with panhypophysitis, and 8 with infundibuloneurohypophysitis. The frequency of secondary hypoadrenalism was similar in adenohypophysitis, panhypophysitis, and infundibuloneurohypophysitis. Growth hormone deficit and secondary hypogonadism occurred more frequently in infundibuloneurohypophysitis than in adenohypophysitis and panhypophysitis (p = 0.009; p = 0.04). All cases of multiple pituitary secretion deficits occurred in cases of infundibuloneurohypophysitis (p = 0.04). No correlations between hypophysitis subtype and anti-pituitary and anti-hypothalamus autoantibodies were found. A higher frequency of extractable nuclear antigens (ENA) and anti-nuclear antibodies (ANA) was found in cases of panhypophysitis (OR 5.0, 95% CI 0.86-28.8, p < 0.001, and OR 1.8, 95% CI 1.1-3.2, p = 0.02, respectively) as compared to adenohypophysitis and infundibuloneurohypophysitis. CONCLUSION: Infundibuloneurohypophysitis should be taken into account in the etiological diagnosis of hypopituitarism, particularly if it is associated with diabetes insipidus and in cases of growth hormone deficit, secondary hypogonadism, or multiple hormone deficits. Contrast-enhanced MRI is crucial for the clinical and noninvasive diagnosis of hypophysitis. Screening for autoantibodies, particularly anti-ENA and anti-ANA, is strongly suggested in the clinical context of hypophysitis.

Persistent Urogenital Sinus: Diagnostic Imaging for Clinical Management. What Does the Radiologist Need to Know?

BACKGROUND: Persistent urogenital sinus (PUGS) is a congenital pathological condition characterized by an abnormal communication between the urethra and vagina. It may be a part of a complex syndrome and can be more often associated with congenital malformations affecting the genitourinary tract system (33%) such as intersex, rectovaginal communication, bladder agenesis, absence of vagina, and hydrocolpos. The correct radiological assessment of PUGS is especially useful for clinicians since the exact anatomical evaluation of this abnormality is a crucial factor for surgical planning. The imaging study modalities, which are essentially based on ultrasonography, voiding cystourethrography, and magnetic resonance imaging, could be misinterpreted if not correctly performed. AIM: The aim of this article is to highlight this rare pathological condition and to help general radiologists in achieving the correct technical approach for the diagnosis. Special attention will be paid in discussing the role of different imaging modalities and their contribution to the diagnosis and clinical management of patients.

CT-based radiogenomics of intrahepatic cholangiocarcinoma.

BACKGROUND: Intrahepatic cholangiocarcinoma (ICC) is an aggressive disease with increasing incidence and its genetic alterations could be the target of systemic therapies. AIMS: To elucidate if radiomics extracted from computed tomography (CT) may non-invasively predict ICC genetic alterations. METHODS: All consecutive patients with a diagnosis of a mass-forming ICC (01/2016-06/2022) were considered. Inclusion criteria were availability of a high-quality contrast-enhanced CT and molecular profiling by NGS or FISH for FGFR2 fusion/rearrangement. The CT scan at diagnosis was considered. Genetic analyses were performed on surgical specimens (resectable patients) or biopsies (unresectable ones). The radiomic features were extracted using the LifeX software. Multivariate predictive models of the commonest genetic alterations were built. RESULTS: In the 90 enrolled patients (58 NGS/32 FISH, median age 65 years), the most common genetic alterations were FGFR2 (20/90), IDH1 (10/58), and KRAS (9/58). At internal validation, the combined clinical-radiomic models achieved the best performance for the prediction of FGFR2 (AUC = 0.892) and IDH1 status (AUC = 0.819), outperforming the pure clinical and radiomic models. The radiomic model for predicting KRAS mutations achieved an AUC = 0.767 (vs. 0.660 of the clinical model) without further improvements with the addition of clinical features. CONCLUSIONS: CT-based radiomics provides a reliable non-invasive prediction of ICC genetic status with a major impact on therapeutic strategies.

Accuracy of Radiomics in Predicting IDH Mutation Status in Diffuse Gliomas: A Bivariate Meta-Analysis.

Purpose To perform a systematic review and meta-analysis assessing the predictive accuracy of radiomics in the noninvasive determination of isocitrate dehydrogenase (IDH) status in grade 4 and lower-grade diffuse gliomas. Materials and Methods A systematic search was performed in the PubMed, Scopus, Embase, Web of Science, and Cochrane Library databases for relevant articles published between January 1, 2010, and July 7, 2021. Pooled sensitivity and specificity across studies were estimated. Risk of bias was evaluated using Quality Assessment of Diagnostic Accuracy Studies-2, and methods were evaluated using the radiomics quality score (RQS). Additional subgroup analyses were performed according to tumor grade, RQS, and number of sequences used (PROSPERO ID: CRD42021268958). Results Twenty-six studies that included 3280 patients were included for analysis. The pooled sensitivity and specificity of radiomics for the detection of IDH mutation were 79% (95% CI: 76, 83) and 80% (95% CI: 76, 83), respectively. Low RQS scores were found overall for the included works. Subgroup analyses showed lower false-positive rates in very low RQS studies (RQS < 6) (meta-regression, z = -1.9; P = .02) compared with adequate RQS studies. No substantial differences were found in pooled sensitivity and specificity for the pure grade 4 gliomas group compared with the all-grade gliomas group (81% and 86% vs 79% and 79%, respectively) and for studies using single versus multiple sequences (80% and 77% vs 79% and 82%, respectively). Conclusion The pooled data showed that radiomics achieved good accuracy performance in distinguishing IDH mutation status in patients with grade 4 and lower-grade diffuse gliomas. The overall methodologic quality (RQS) was low and introduced potential bias. Keywords: Neuro-Oncology, Radiomics, Integration, Application Domain, Glioblastoma, IDH Mutation, Radiomics Quality Scoring Supplemental material is available for this article. Published under a CC BY 4.0 license.

Advanced Imaging Supports the Mechanistic Role of Autoimmunity and Plaque Rupture in COVID-19 Heart Involvement.

The cardiovascular system is frequently affected by coronavirus disease-19 (COVID-19), particularly in hospitalized cases, and these manifestations are associated with a worse prognosis. Most commonly, heart involvement is represented by myocarditis, myocardial infarction, and pulmonary embolism, while arrhythmias, heart valve damage, and pericarditis are less frequent. While the clinical suspicion is necessary for a prompt disease recognition, imaging allows the early detection of cardiovascular complications in patients with COVID-19. The combination of cardiothoracic approaches has been proposed for advanced imaging techniques, i.e., CT scan and MRI, for a simultaneous evaluation of cardiovascular structures, pulmonary arteries, and lung parenchyma. Several mechanisms have been proposed to explain the cardiovascular injury, and among these, it is established that the host immune system is responsible for the aberrant response characterizing severe COVID-19 and inducing organ-specific injury. We illustrate novel evidence to support the hypothesis that molecular mimicry may be the immunological mechanism for myocarditis in COVID-19. The present article provides a comprehensive review of the available evidence of the immune mechanisms of the COVID-19 cardiovascular injury and the imaging tools to be used in the diagnostic workup. As some of these techniques cannot be implemented for general screening of all cases, we critically discuss the need to maximize the sustainability and the specificity of the proposed tests while illustrating the findings of some paradigmatic cases.

A virtual biopsy of liver parenchyma to predict the outcome of liver resection.

The preoperative risk assessment of liver resections (LR) is still an open issue. Liver parenchyma characteristics influence the outcome but cannot be adequately evaluated in the preoperative setting. The present study aims to elucidate the contribution of the radiomic analysis of non-tumoral parenchyma to the prediction of complications after elective LR. All consecutive patients undergoing LR between 2017 and 2021 having a preoperative computed tomography (CT) were included. Patients with associated biliary/colorectal resection were excluded. Radiomic features were extracted from a virtual biopsy of non-tumoral liver parenchyma (a 2 mL cylinder) outlined in the portal phase of preoperative CT. Data were internally validated. Overall, 378 patients were analyzed (245 males/133 females-median age 67 years-39 cirrhotics). Radiomics increased the performances of the preoperative clinical models for both liver dysfunction (at internal validaton, AUC = 0.727 vs. 0.678) and bile leak (AUC = 0.744 vs. 0.614). The final predictive model combined clinical and radiomic variables: for bile leak, segment 1 resection, exposure of Glissonean pedicles, HU-related indices, NGLDM_Contrast, GLRLM indices, and GLZLM_ZLNU; for liver dysfunction, cirrhosis, liver function tests, major hepatectomy, segment 1 resection, and NGLDM_Contrast. The combined clinical-radiomic model for bile leak based on preoperative data performed even better than the model including the intraoperative data (AUC = 0.629). The textural features extracted from a virtual biopsy of non-tumoral liver parenchyma improved the prediction of postoperative liver dysfunction and bile leak, implementing information given by standard clinical data. Radiomics should become part of the preoperative assessment of candidates to LR.

Generative Adversarial Networks in Brain Imaging: A Narrative Review.

Artificial intelligence (AI) is expected to have a major effect on radiology as it demonstrated remarkable progress in many clinical tasks, mostly regarding the detection, segmentation, classification, monitoring, and prediction of diseases. Generative Adversarial Networks have been proposed as one of the most exciting applications of deep learning in radiology. GANs are a new approach to deep learning that leverages adversarial learning to tackle a wide array of computer vision challenges. Brain radiology was one of the first fields where GANs found their application. In neuroradiology, indeed, GANs open unexplored scenarios, allowing new processes such as image-to-image and cross-modality synthesis, image reconstruction, image segmentation, image synthesis, data augmentation, disease progression models, and brain decoding. In this narrative review, we will provide an introduction to GANs in brain imaging, discussing the clinical potential of GANs, future clinical applications, as well as pitfalls that radiologists should be aware of.

The added value of artificial intelligence to LI-RADS categorization: A systematic review.

PURPOSE: The objective of this systematic review was to critically assess the available literature on deep learning (DL) and radiomics applied to the Liver Imaging Reporting and Data System (LI-RADS) in terms of 1) automatic LI-RADS classification of liver nodules; 2) the contribution of DL and radiomics to human evaluation in the classification of liver nodules following LI-RADS protocol. MATERIALS AND METHODS: A literature search was conducted to identify original research studies published up to April 2021. The inclusion criteria were: English language, focus on computed tomography (CT) and/or magnetic resonance (MR) with specified number of patients and lesions, adoption of LI-RADS classification for the detected hepatic lesions, and application of AI in the classification of liver nodules. Review articles, conference papers, editorials and commentaries, animal studies or studies with absence of AI and/or LI-RADS were excluded. After screening 221 articles, 11 studies were included in this review. RESULTS: All the included studies proved that DL and radiomics have high performances in liver nodules classification, sometimes similar or better than human evaluation. The best performances of DL was an AUC of 0.95 on MR and the best performance of radiomics was AUC of 0.98 either on CT and MR, while the lower ones were respectively AUC of 0.63 either on CT and MR for DL and AUC of 0.70 on CT for radiomics. CONCLUSION: DL and radiomics could be a useful tool in assisting radiologists in the diagnosis and classification of liver nodules according to LI-RADS.

Artificial Intelligence Applied to Pancreatic Imaging: A Narrative Review.

The diagnosis, evaluation, and treatment planning of pancreatic pathologies usually require the combined use of different imaging modalities, mainly, computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET). Artificial intelligence (AI) has the potential to transform the clinical practice of medical imaging and has been applied to various radiological techniques for different purposes, such as segmentation, lesion detection, characterization, risk stratification, or prediction of response to treatments. The aim of the present narrative review is to assess the available literature on the role of AI applied to pancreatic imaging. Up to now, the use of computer-aided diagnosis (CAD) and radiomics in pancreatic imaging has proven to be useful for both non-oncological and oncological purposes and represents a promising tool for personalized approaches to patients. Although great developments have occurred in recent years, it is important to address the obstacles that still need to be overcome before these technologies can be implemented into our clinical routine, mainly considering the heterogeneity among studies.

An individualized algorithm to predict mortality in COVID-19 pneumonia: a machine learning based study.

Introduction: Identifying SARS-CoV-2 patients at higher risk of mortality is crucial in the management of a pandemic. Artificial intelligence techniques allow one to analyze large amounts of data to find hidden patterns. We aimed to develop and validate a mortality score at admission for COVID-19 based on high-level machine learning. Material and methods: We conducted a retrospective cohort study on hospitalized adult COVID-19 patients between March and December 2020. The primary outcome was in-hospital mortality. A machine learning approach based on vital parameters, laboratory values and demographic features was applied to develop different models. Then, a feature importance analysis was performed to reduce the number of variables included in the model, to develop a risk score with good overall performance, that was finally evaluated in terms of discrimination and calibration capabilities. All results underwent cross-validation. Results: 1,135 consecutive patients (median age 70 years, 64% male) were enrolled, 48 patients were excluded, and the cohort was randomly divided into training (760) and test (327) groups. During hospitalization, 251 (22%) patients died. After feature selection, the best performing classifier was random forest (AUC 0.88 ±0.03). Based on the relative importance of each variable, a pragmatic score was developed, showing good performances (AUC 0.85 ±0.025), and three levels were defined that correlated well with in-hospital mortality. Conclusions: Machine learning techniques were applied in order to develop an accurate in-hospital mortality risk score for COVID-19 based on ten variables. The application of the proposed score has utility in clinical settings to guide the management and prognostication of COVID-19 patients.

Artificial intelligence processing electronic health records to identify commonalities and comorbidities cluster at Immuno Center Humanitas.

Background: Comorbidities are common in chronic inflammatory conditions, requiring multidisciplinary treatment approach. Understanding the link between a single disease and its comorbidities is important for appropriate treatment and management. We evaluate the ability of an NLP-based process for knowledge discovery to detect information about pathologies, patients' phenotype, doctors' prescriptions and commonalities in electronic medical records, by extracting information from free narrative text written by clinicians during medical visits, resulting in the extraction of valuable information and enriching real world evidence data from a multidisciplinary setting. Methods: We collected clinical notes from the Allergy Department of Humanitas Research Hospital written in the last 3 years and used it to look for diseases that cluster together as comorbidities associated to the main pathology of our patients, and for the extent of prescription of systemic corticosteroids, thus evaluating the ability of NLP-based tools for knowledge discovery to extract structured information from free text. Results: We found that the 3 most frequent comorbidities to appear in our clusters were asthma, rhinitis, and urticaria, and that 991 (of 2057) patients suffered from at least one of these comorbidities. The clusters which co-occur particularly often are oral allergy syndrome and urticaria (131 patients), angioedema and urticaria (105 patients), rhinitis and asthma (227 patients). With regards to systemic corticosteroid prescription volume by our clinicians, we found it was lower when compared to the therapy the patients followed before coming to our attention, with the exception of two diseases: Chronic obstructive pulmonary disease and Angioedema. Conclusions: This analysis seems to be valid and is confirmed by the data from the literature. This means that NLP tools could have significant role in many other research fields of medicine, as it may help identify other important, and possibly previously neglected clusters of patients with comorbidities and commonalities. Another potential benefit of this approach lies in its potential ability to foster a multidisciplinary approach, using the same drugs to treat pathologies normally treated by physicians in different branches of medicine, thus saving resources and improving the pharmacological management of patients.

Advanced imaging and Crohn's disease: An overview of clinical application and the added value of artificial intelligence.

PURPOSE: The purpose of this narrative review is to describe the clinical applications of advanced computed tomography (CT) and magnetic resonance (MRI) techniques in patients affected by Crohn's disease (CD), giving insights about the added value of artificial intelligence (AI) in this field. METHODS: We performed a literature search comparing standardized and advanced imaging techniques for CD diagnosis. Cross-sectional imaging is essential for the identification of lesions, the assessment of active or relapsing disease and the evaluation of complications. RESULTS: The studies reviewed show that new advanced imaging techniques and new MRI sequences could be integrated into standard protocols, to achieve a reliable quantification of CD activity, improve the lesions' characterization and the evaluation of therapy response. These promising tools are: dual-energy CT (DECT) post-processing techniques, diffusion-weighted MRI (DWI-MRI), dynamic contrast-enhanced MRI (DCE-MRI), Magnetization Transfer MRI (MT-MRI) and CINE-MRI. Furthermore, AI solutions show a potential when applied to radiological techniques in these patients. Machine learning (ML) algorithms and radiomic features prove to be useful in improving the diagnostic accuracy of clinicians and in attempting a personalized medicine approach, stratifying patients by predicting their prognosis. CONCLUSIONS: Advanced imaging is crucial in the diagnosis, lesions' characterisation and in the estimation of the abdominal involvement in CD. New AI developments are promising tools that could support doctors in the management of CD affected patients.

Word2vec Word Embedding-Based Artificial Intelligence Model in the Triage of Patients with Suspected Diagnosis of Major Ischemic Stroke: A Feasibility Study.

BACKGROUND: The possible benefits of using semantic language models in the early diagnosis of major ischemic stroke (MIS) based on artificial intelligence (AI) are still underestimated. The present study strives to assay the feasibility of the word2vec word embedding-based model in decreasing the risk of false negatives during the triage of patients with suspected MIS in the emergency department (ED). METHODS: The main ICD-9 codes related to MIS were used for the 7-year retrospective data collection of patients managed at the ED with a suspected diagnosis of stroke. The data underwent "tokenization" and "lemmatization". The word2vec word-embedding algorithm was used for text data vectorization. RESULTS: Out of 648 MIS, the word2vec algorithm successfully identified 83.9% of them, with an area under the curve of 93.1%. CONCLUSIONS: Natural language processing (NLP)-based models in triage have the potential to improve the early detection of MIS and to actively support the clinical staff.

A machine learning risk model based on preoperative computed tomography scan to predict postoperative outcomes after pancreatoduodenectomy.

Clinically relevant postoperative pancreatic fistula (CR-POPF) is a life-threatening complication following pancreaticoduodenectomy (PD). Individualized preoperative risk assessment could improve clinical management and prevent or mitigate adverse outcomes. The aim of this study is to develop a machine learning risk model to predict occurrence of CR-POPF after PD from preoperative computed tomography (CT) scans. A total of 100 preoperative high-quality CT scans of consecutive patients who underwent pancreaticoduodenectomy in our institution between 2011 and 2019 were analyzed. Radiomic and morphological features extracted from CT scans related to pancreatic anatomy and patient characteristics were included as variables. These data were then assessed by a machine learning classifier to assess the risk of developing CR-POPF. Among the 100 patients evaluated, 20 had CR-POPF. The predictive model based on logistic regression demonstrated specificity of 0.824 (0.133) and sensitivity of 0.571 (0.337), with an AUC of 0.807 (0.155), PPV of 0.468 (0.310) and NPV of 0.890 (0.084). The performance of the model minimally decreased utilizing a random forest approach, with specificity of 0.914 (0.106), sensitivity of 0.424 (0.346), AUC of 0.749 (0.209), PPV of 0.502 (0.414) and NPV of 0.869 (0.076). Interestingly, using the same data, the model was also able to predict postoperative overall complications and a postoperative length of stay over the median with AUCs of 0.690 (0.209) and 0.709 (0.160), respectively. These findings suggest that preoperative CT scans evaluated by machine learning may provide a novel set of information to help clinicians choose a tailored therapeutic pathway in patients candidated to pancreatoduodenectomy.