A Multicenter International Retrospective Investigation Assessing the Prognostic Role of Inflammation-Based Scores (Neutrophil-to-Lymphocyte, Lymphocyte-to-Monocyte, and Platelet-to-Lymphocyte Ratios) in Patients with Intermediate-Stage Hepatocellular Carcinoma (HCC) Undergoing Chemoembolizations of the Liver.

BACKGROUND: The utilization of inflammation-based scores, such as the Neutrophil-to-Lymphocyte Ratio (NLR), Lymphocyte-to-Monocyte Ratio (LMR), and Platelet-to-Lymphocyte Ratio (PLR), has garnered attention for their potential as prognostic indicators in various cancers. However, their predictive role in patients with intermediate-stage HCC undergoing transcatheter arterial chemoembolization (TACE) remains an area that requires further investigation, as early recognition of TACE refractoriness holds the potential to guide tailored therapeutic interventions. METHODS: This multicenter international retrospective study analyzed data from patients with intermediate-stage HCC undergoing TACE between 2018 and 2024. Inflammation-based scores (NLR, LMR, PLR) were assessed preoperatively to predict treatment outcomes. RESULTS: Two hundred and fourteen patients were enrolled. Preoperative LMR showed the largest area under the curve for the prediction of 6-months PFS, based on the ROC curve analysis. Both high LMR (≥2.24) and low NLR (<4.72) were associated with improved objective response rates and 6-month progression-free survival. Lymphocyte count emerged as a strong predictor of treatment response in both simple (p < 0.001) and multiple (p < 0.001) logistic regression analyses. CONCLUSIONS: This study highlights the prognostic value of inflammation-based scores, particularly LMR and NLR, in predicting the treatment response and short-term outcomes of patients with intermediate-stage HCC undergoing TACE. Future investigations should focus on validating these scores' clinical applicability and assessing their impact on long-term patient survival and therapeutic decision-making.

Quantitative ultrasound (QUS) in the evaluation of liver steatosis: data reliability in different respiratory phases and body positions.

Liver steatosis is the most common chronic liver disease and affects 10-24% of the general population. As the grade of disease can range from fat infiltration to steatohepatitis and cirrhosis, an early diagnosis is needed to set the most appropriate therapy. Innovative noninvasive radiological techniques have been developed through MRI and US. MRI-PDFF is the reference standard, but it is not so widely diffused due to its cost. For this reason, ultrasound tools have been validated to study liver parenchyma. The qualitative assessment of the brightness of liver parenchyma has now been supported by quantitative values of attenuation and scattering to make the analysis objective and reproducible. We aim to demonstrate the reliability of quantitative ultrasound in assessing liver fat and to confirm the inter-operator reliability in different respiratory phases. We enrolled 45 patients examined during normal breathing at rest, peak inspiration, peak expiration, and semi-sitting position. The highest inter-operator agreement in both attenuation and scattering parameters was achieved at peak inspiration and peak expiration, followed by semi-sitting position. In conclusion, this technology also allows to monitor uncompliant patients, as it grants high reliability and reproducibility in different body position and respiratory phases.

Transcatheter Arterial Embolization for Bleeding Related to Pelvic Trauma: Comparison of Technical and Clinical Results between Hemodynamically Stable and Unstable Patients.

BACKGROUND: Endovascular intervention is now the primary line of therapy for arterial injury brought on by pelvic trauma since it can significantly reduce considerable morbidity associated with surgery and can swiftly access and control bleeding sites. Despite international guidelines and widespread awareness of the role of angioembolization in clinical practice, robust evidence comparing the outcomes of angioembolization in hemodynamically stable and unstable patients is still lacking. This study aims to directly compare the outcomes of angioembolization for the treatment of pelvic traumatic arterial injury in patients with hemodynamic stability vs. hemodynamic instability. METHODS: In our multicenter retrospective investigation, we analyzed data from consecutive patients who underwent, from January 2020 to May 2023, angioembolization for traumatic pelvic arterial injury. RESULTS: In total, 116 angioembolizations were performed. Gelatin sponges (56.9%) and coils (25.9%) were the most widely used embolic agents. The technical and clinical success rates were 100% and 91.4%, respectively. No statistically significant differences were observed between the two groups in terms of technical success, clinical success, procedure-related complication rate, or 30-day bleeding-related mortality. CONCLUSIONS: Angioembolization is an effective and safe option for the management of traumatic pelvic arterial lesions even in hemodynamically unstable patients, despite technical variations such as greater use of prophylactic angioembolization.

Explainable Convolutional Neural Networks for Brain Cancer Detection and Localisation.

Brain cancer is widely recognised as one of the most aggressive types of tumors. In fact, approximately 70% of patients diagnosed with this malignant cancer do not survive. In this paper, we propose a method aimed to detect and localise brain cancer, starting from the analysis of magnetic resonance images. The proposed method exploits deep learning, in particular convolutional neural networks and class activation mapping, in order to provide explainability by highlighting the areas of the medical image related to brain cancer (from the model point of view). We evaluate the proposed method with 3000 magnetic resonances using a free available dataset. The results we obtained are encouraging. We reach an accuracy ranging from 97.83% to 99.67% in brain cancer detection by exploiting four different models: VGG16, ResNet50, Alex_Net, and MobileNet, thus showing the effectiveness of the proposed method.

Endovascular Treatment of Visceral Artery Pseudoaneurysms with Ethylene-Vinyl Alcohol (EVOH) Copolymer-Based Non-Adhesive Liquid Embolic Agents (NALEAs).

Background and Objectives: Treatment of visceral artery pseudoaneurysms (VAPs) is always indicated regardless of their diameters, as their risk of rupture is significantly higher than that of visceral artery aneurysms. The invasiveness of surgery and its associated complications have led to a shift in favor of radiological interventions as the initial treatment of choice. However, there are still some unanswered questions on endovascular treatment of VAPs regarding the optimal endovascular technique and the efficacy and safety outcomes. The purpose of this multicenter study was to retrospectively evaluate the effectiveness and safety of endovascular treatment of visceral pseudoaneurysms using Ethylene-Vinyl Alcohol (EVOH) Copolymer-Based Non-Adhesive Liquid Embolic Agents (NALEAs). Materials and Methods: Consecutive patients who underwent endovascular embolization with EVOH-based NALEAs for visceral artery pseudoaneurysms between January 2018 and June 2023 were retrospectively evaluated. Results: 38 embolizations were performed. Technical success was achieved in all patients. The clinical success rate was high (92.1% overall), with no significant differences between ruptured and unruptured VAPs (p = 0.679). Seven patients (18.4%) experienced procedure-related complications, related to one case of non-target embolization, four splenic abscesses due to end-organ infarction, and two femoral pseudoaneurysms. The rates of procedure-related complications, end-organ infarction, and vascular access-site complications did not significantly differ between ruptured and unruptured VAPs (p > 0.05). Conclusions: Both ruptured and unruptured visceral pseudoaneurysms can be effectively and safely treated with NALEA-based endovascular embolization. We suggest considering the use of NALEAs, particularly in specific clinical cases that highlight their advantages, including patients with coagulopathy, fragile vessels, and embolization targets that are located at a considerable distance from the microcatheter tip and are otherwise difficult to reach.

Transcatheter Arterial Embolization (TAE) of Cancer-Related Bleeding.

Background and Objectives: Roughly 10% of cancer patients experience an episode of bleeding. The bleeding severity can range from occasional trivial bleeds to major bleeding. The treatment for the bleeding may vary, depending on the clinical condition and anatomical site, and may include various strategies, among which TAE is a cornerstone of major bleeding management. However, the existing literature on tumor hemorrhages is inconsistent. The objective of this multicenter retrospective cohort study was to evaluate the effectiveness and safety of arterial embolization in the treatment of tumor hemorrhages in patients with solid cancers. Materials and Methods: The data for patients with solid cancers undergoing TAE for the management of tumor hemorrhages from January 2020 to May 2023 were gathered. Results: A total of 92 patients with cancer-related bleeding were treated between January 2020 and May 2023. No bleeding was detected by X-ray angiography (XA) in 12 (13%) cases; therefore, a blind embolization was performed. The most common bleeding site was the liver (21.7%). A total of 66 tumor hemorrhages were spontaneous. The most commonly used embolic agent was polyvinyl alcohol (PVA) particles (30.4%). Technical success was achieved in 82 (89.1%) cases, with an 84.8% clinical success rate related to 14 cases of rebleeding. Proximal embolization was performed for 19 (20.7%) patients. Complications were recorded for 10 (10.9%) patients. The 30-day bleeding-related mortality was 15.2%. The technical success, clinical success, proximal embolization rate, and 30-day rebleeding were worse in the subset of patients undergoing TAE with coils. Conclusions: Transcatheter arterial embolization (TAE) represents a viable and potentially life-saving therapeutic approach in the management of tumor hemorrhages, demonstrating a notable effectiveness and safety. The TAE of bleeding tumors using coils resulted in a higher rate of non-superselective proximal embolization, with a trend toward lower clinical success rates and higher rebleeding episodes.

A Multicenter Retrospective Cohort Study Evaluating the Clinical Outcomes of Patients with Coagulopathy Undergoing Transcatheter Arterial Embolization (TAE) for Acute Non-Neurovascular Bleeding.

Background and Objectives: Transcatheter arterial embolization (TAE) is the mainstay of treatment for acute major hemorrhage, even in patients with coagulopathy and spontaneous bleeding. Coagulopathy is associated with worsening bleeding severity and higher mortality and clinical failure rates. Furthermore, some unanswered questions remain, such as the definition of coagulopathy, the indication for TAE or conservative treatment, and the choice of embolic agent. This study aims to assess the efficacy and safety of TAE for spontaneous non-neurovascular acute bleeding in patients with coagulopathy. Materials and Methods: This study is a multicenter analysis of retrospectively collected data of consecutive patients with coagulopathy who had undergone, from January 2018 to May 2023, transcatheter arterial embolization for the management of spontaneous hemorrhages. Results: During the study interval (January 2018-May 2023), 120 patients with coagulopathy underwent TAE for spontaneous non-neurovascular acute bleeding. The abdominal wall was the most common bleeding site (72.5%). The most commonly used embolic agent was polyvinyl alcohol (PVA) particles or microspheres (25.0%), whereas coils and gelatin sponge together accounted for 32.5% of the embolic agents used. Technical success was achieved in all cases, with a 92.5% clinical success rate related to 9 cases of rebleeding. Complications were recorded in 12 (10%) patients. Clinical success was significantly better in the group of patients who underwent correction of the coagulopathy within 24 h of TAE. Conclusions: Transcatheter arterial embolization (TAE) is effective and safe for the management of acute non-neurovascular bleeding in patients with coagulopathy. Correction of coagulopathy should not delay TAE and vice versa, as better clinical outcomes were noted in the subgroup of patients undergoing correction of coagulopathy within 24 h of TAE.

Increasing differential diagnosis between lipoma and liposarcoma through radiomics: a narrative review.

Soft tissue sarcomas (STSs) are rare, heterogeneous, and very often asymptomatic diseases. Their diagnosis is fundamental, as is the identification of the degree of malignancy, which may be high, medium, or low. The Italian Medical Oncology Association and European Society of Medical Oncology (ESMO) guidelines recommend magnetic resonance imaging (MRI) because the clinical examination is typically ineffective. The diagnosis of these rare diseases with artificial intelligence (AI) techniques presents reduced datasets and therefore less robust methods. However, the combination of AI techniques with radiomics may be a new angle in diagnosing rare diseases such as STSs. Results obtained are promising within the literature, not only for the performance but also for the explicability of the data. In fact, one can make tumor classification, site localization, and prediction of the risk of developing metastasis. Thanks to the synergy between computer scientists and radiologists, linking numerical features to radiological evidence with excellent performance could be a new step forward for the diagnosis of rare diseases.

Peritoneal Carcinosis: What the Radiologist Needs to Know.

Peritoneal carcinosis is a condition characterized by the spread of cancer cells to the peritoneum, which is the thin membrane that lines the abdominal cavity. It is a serious condition that can result from many different types of cancer, including ovarian, colon, stomach, pancreatic, and appendix cancer. The diagnosis and quantification of lesions in peritoneal carcinosis are critical in the management of patients with the condition, and imaging plays a central role in this process. Radiologists play a vital role in the multidisciplinary management of patients with peritoneal carcinosis. They need to have a thorough understanding of the pathophysiology of the condition, the underlying neoplasms, and the typical imaging findings. In addition, they need to be aware of the differential diagnoses and the advantages and disadvantages of the various imaging methods available. Imaging plays a central role in the diagnosis and quantification of lesions, and radiologists play a critical role in this process. Ultrasound, computed tomography, magnetic resonance, and PET/CT scans are used to diagnose peritoneal carcinosis. Each imaging procedure has advantages and disadvantages, and particular imaging techniques are recommended based on patient conditions. Our aim is to provide knowledge to radiologists regarding appropriate techniques, imaging findings, differential diagnoses, and treatment options. With the advent of AI in oncology, the future of precision medicine appears promising, and the interconnection between structured reporting and AI is likely to improve diagnostic accuracy and treatment outcomes for patients with peritoneal carcinosis.

Transcatheter Arterial Embolization (TAE) in the Management of Bleeding in the COVID-19 Patient.

Background and Objectives: Increasing attention is being paid to the coagulation disorders associated with SARS-CoV-2 infection. Bleeding accounts for 3-6% of COVID-19 patient deaths, and is often a forgotten part of the disease. The bleeding risk is enhanced by several factors, including spontaneous heparin-induced thrombocytopenia, thrombocytopenia, the hyperfibrinolytic state, the consumption of coagulation factors, and thromboprophylaxis with anticoagulants. This study aims to assess the efficacy and safety of TAE in the management of bleeding in COVID-19 patients. Materials and Methods: This multicenter retrospective study analyzes data from COVID-19 patients subjected to transcatheter arterial embolization for the management of bleeding from February 2020 to January 2023. Results: Transcatheter arterial embolization was performed in 73 COVID-19 patients for acute non-neurovascular bleeding during the study interval (February 2020-January 2023). Coagulopathy was observed in forty-four (60.3%) patients. The primary cause of bleeding was spontaneous soft tissue hematoma (63%). A 100% technical success rate was recorded; six cases of rebleeding resulted in a 91.8% clinical success rate. No cases of non-target embolization were observed. Complications were recorded in 13 (17.8%) patients. The efficacy and safety endpoints did not differ significantly between the coagulopathy and non-coagulopathy groups. Conclusions: Transcatheter Arterial Embolization (TAE) is an effective, safe and potentially life-saving option for the management of acute non-neurovascular bleeding in COVID-19 patients. This approach is effective and safe even in the subgroup of COVID-19 patients with coagulopathy.

Predicting risk of metastases and recurrence in soft-tissue sarcomas via Radiomics and Formal Methods.

Objective: Soft-tissue sarcomas (STSs) of the extremities are a group of malignancies arising from the mesenchymal cells that may develop distant metastases or local recurrence. In this article, we propose a novel methodology aimed to predict metastases and recurrence risk in patients with these malignancies by evaluating magnetic resonance radiomic features that will be formally verified through formal logic models. Materials and Methods: This is a retrospective study based on a public dataset evaluating MRI scans T2-weighted fat-saturated or short tau inversion recovery and patients having "metastases/local recurrence" (group B) or "no metastases/no local recurrence" (group A) as clinical outcomes. Once radiomic features are extracted, they are included in formal models, on which is automatically verified the logic property written by a radiologist and his computer scientists coworkers. Results: Evaluating the Formal Methods efficacy in predicting distant metastases/local recurrence in STSs (group A vs group B), our methodology showed a sensitivity and specificity of 0.81 and 0.67, respectively; this suggests that radiomics and formal verification may be useful in predicting future metastases or local recurrence development in soft tissue sarcoma. Discussion: Authors discussed about the literature to consider Formal Methods as a valid alternative to other Artificial Intelligence techniques. Conclusions: An innovative and noninvasive rigourous methodology can be significant in predicting local recurrence and metastases development in STSs. Future works can be the assessment on multicentric studies to extract objective disease information, enriching the connection between the radiomic quantitative analysis and the radiological clinical evidences.

Radiomics and Artificial Intelligence Can Predict Malignancy of Solitary Pulmonary Nodules in the Elderly.

Solitary pulmonary nodules (SPNs) are a diagnostic and therapeutic challenge for thoracic surgeons. Although such lesions are usually benign, the risk of malignancy remains significant, particularly in elderly patients, who represent a large segment of the affected population. Surgical treatment in this subset, which usually presents several comorbidities, requires careful evaluation, especially when pre-operative biopsy is not feasible and comorbidities may jeopardize the outcome. Radiomics and artificial intelligence (AI) are progressively being applied in predicting malignancy in suspicious nodules and assisting the decision-making process. In this study, we analyzed features of the radiomic images of 71 patients with SPN aged more than 75 years (median 79, IQR 76-81) who had undergone upfront pulmonary resection based on CT and PET-CT findings. Three different machine learning algorithms were applied-functional tree, Rep Tree and J48. Histology was malignant in 64.8% of nodules and the best predictive value was achieved by the J48 model (AUC 0.9). The use of AI analysis of radiomic features may be applied to the decision-making process in elderly frail patients with suspicious SPNs to minimize the false positive rate and reduce the incidence of unnecessary surgery.

Coronavirus covid-19 detection by means of explainable deep learning.

The coronavirus is caused by the infection of the SARS-CoV-2 virus: it represents a complex and new condition, considering that until the end of December 2019 this virus was totally unknown to the international scientific community. The clinical management of patients with the coronavirus disease has undergone an evolution over the months, thanks to the increasing knowledge of the virus, symptoms and efficacy of the various therapies. Currently, however, there is no specific therapy for SARS-CoV-2 virus, know also as Coronavirus disease 19, and treatment is based on the symptoms of the patient taking into account the overall clinical picture. Furthermore, the test to identify whether a patient is affected by the virus is generally performed on sputum and the result is generally available within a few hours or days. Researches previously found that the biomedical imaging analysis is able to show signs of pneumonia. For this reason in this paper, with the aim of providing a fully automatic and faster diagnosis, we design and implement a method adopting deep learning for the novel coronavirus disease detection, starting from computed tomography medical images. The proposed approach is aimed to detect whether a computed tomography medical images is related to an healthy patient, to a patient with a pulmonary disease or to a patient affected with Coronavirus disease 19. In case the patient is marked by the proposed method as affected by the Coronavirus disease 19, the areas symptomatic of the Coronavirus disease 19 infection are automatically highlighted in the computed tomography medical images. We perform an experimental analysis to empirically demonstrate the effectiveness of the proposed approach, by considering medical images belonging from different institutions, with an average time for Coronavirus disease 19 detection of approximately 8.9 s and an accuracy equal to 0.95.

The role of Lung Ultrasound in the diagnosis of SARS-COV-2 disease in pregnant women.

AIM: To evaluate the role of lung ultrasound (LUS) in recognizing lung abnormalities in pregnant women affected by COVID-19 pneumonia. MATERIALS AND METHODS: An observational study analyzing LUS patterns in 60 consecutively enrolled pregnant women affected by COVID-19 infection was performed. LUS was performed by using a standardized protocol by Soldati et al. The scoring system of LUS findings ranged from 0 to 3 in increasing alteration severity. The highest score obtained from each landmark was reported and the sum of the 12 zones examined was calculated. RESULTS: Patients were divided into two groups: 26 (43.3%) patients with respiratory symptoms and 32 (53.3%) patients without respiratory symptoms; 2 patients were asymptomatic (3.3%). Among the patients with respiratory symptoms 3 (12.5%) had dyspnea that required a mild Oxygen therapy. A significant correlation was found between respiratory symptoms and LUS score (p < 0.001) and between gestational weeks and respiratory symptoms (p = 0.023). Regression analysis showed that age and respiratory symptoms were risk factors for highest LUS score (p < 0.005). DISCUSSION: LUS can affect the clinical decision course and can help in stratifying patients according to its findings. The lack of ionizing radiation and its repeatability makes it a reliable diagnostic tool in the management of pregnant women.

Glenoid bone loss in anterior shoulder dislocation: a multicentric study to assess the most reliable imaging method.

PURPOSE: The aim of this multicentric study was to assess which imaging method has the best inter-reader agreement for glenoid bone loss quantification in anterior shoulder instability. A further aim was to calculate the inter-method agreement comparing bilateral CT with unilateral CT and MR arthrography (MRA) with CT measurements. Finally, calculations were carried out to find the least time-consuming method. METHOD: A retrospective evaluation was performed by 9 readers (or pairs of readers) on a consecutive series of 110 patients with MRA and bilateral shoulder CT. Each reader was asked to calculate the glenoid bone loss of all patients using the following methods: best fit circle area on both MRA and CT images, maximum transverse glenoid width on MRA and CT, CT PICO technique, ratio of the maximum glenoid width to height on MRA and CT, and length of flattening of the anterior glenoid curvature on MRA and CT. Using Pearson's correlation coefficient (PCC), the following agreement values were calculated: the inter-reader for each method, the inter-method for MRA with CT quantifications and the inter-method for CT best-fit circle area and CT PICO. Statistical analysis was carried out to compare the time employed by the readers for each method. RESULTS: Inter-reader agreement PCC mean values were the following: 0.70 for MRA and 0.77 for CT using best fit circle diameter, 0.68 for MRA and 0.72 for CT using best fit circle area, 0.75 for CT PICO, 0.64 for MRA and 0.62 for CT anterior straight line and 0.49 for MRA and 0.43 for CT using length-to-width ratio. CT-MRA inter-modality PCC mean values were 0.9 for best fit circle diameter, 0.9 for best fit circle area, 0.62 for anterior straight line and 0.94 for length-to-width methods. PCC mean value comparing unilateral CT with PICO CT methods was 0.8. MRA best fit circle area method was significantly faster than the same method performed on CT (p = 0.031), while no significant difference was seen between CT and MRA for remaining measurements. CONCLUSIONS: CT PICO is the most reliable imaging method, but both CT and MRA can be reliably used to assess glenoid bone loss. Best fit circle area CT and MRA methods are valuable alternative measurement techniques.

The Role of Ultrasound in Chronic Intestinal Diseases in Pediatric Patients.

Chronic inflammatory bowel diseases (IBD) are chronic disorders of the gastrointestinal tract, with an increasing incidence in pediatric populations. Ultrasound of the intestinal wall represents the first-line imaging technique in children since it is a noninvasive method, is free of ionizing radiation, and is inexpensive. Furthermore, the absence of intestinal wall thickening has a good negative predictive value for IBD, which is greater for Crohn's disease than for ulcerative colitis. Ultrasound is used for the diagnosis of disease, for the differential diagnosis in IBD, in the follow-up of known IBD, in the definition of the site and extent of the disease, for the diagnosis of intestinal complications, for the evaluation of disease activity, in the definition of prognostic parameters, and in the post-operative follow-up.

Explainability of radiomics through formal methods.

BACKGROUND AND OBJECTIVE: Artificial Intelligence has proven to be effective in radiomics. The main problem in using Artificial Intelligence is that researchers and practitioners are not able to know how the predictions are generated. This is currently an open issue because results' explainability is advantageous in understanding the reasoning behind the model, both for patients than for implementing a feedback mechanism for medical specialists using decision support systems. METHODS: Addressing transparency issues related to the Artificial Intelligence field, the innovative technique of Formal methods use a mathematical logic reasoning to produce an automatic, quick and reliable diagnosis. In this paper we analyze results given by the adoption of Formal methods for the diagnosis of the Coronavirus disease: specifically, we want to analyse and understand, in a more medical way, the meaning of some radiomic features to connect them with clinical or radiological evidences. RESULTS: In particular, the usage of Formal methods allows the authors to do statistical analysis on the feature value distributions, to do pattern recognition on disease models, to generalize the model of a disease and to reach high performances of results and interpretation of them. A further step for explainability can be accounted by the localization and selection of the most important slices in a multi-slice approach. CONCLUSIONS: In conclusion, we confirmed the clinical significance of some First order features as Skewness and Kurtosis. On the other hand, we suggest to decline the use of the Minimum feature because of its intrinsic connection with the Computational Tomography exam of the lung.

A novel methodology for head and neck carcinoma treatment stage detection by means of model checking.

CONTEXT: Head and neck cancers are diagnosed at an annual rate of 3% to 7% with respect to the total number of cancers, and 50% to 75% of such new tumours occur in the upper aerodigestive tract. PURPOSE: In this paper we propose formal methods based approach aimed to identify the head and neck tumour treatment stage by means of model checking. We exploit a set of radiomic features to model medical imaging as a labelled transition system to verify treatment stage properties. MAIN FINDINGS: We experiment the proposed method using a public dataset related to computed tomography images obtained in different treatment stages, reaching an accuracy ranging from 0.924 to 0.978 in treatment stage detection. PRINCIPAL CONCLUSIONS: The study confirms the effectiveness of the adoption of formal methods in the head and neck carcinoma treatment stage detection to support radiologists and pathologists.

Contrast MR-Based Radiomics and Machine Learning Analysis to Assess Clinical Outcomes following Liver Resection in Colorectal Liver Metastases: A Preliminary Study.

PURPOSE: To assess radiomics features efficacy obtained by arterial and portal MRI phase in the prediction of clinical outcomes in the colorectal liver metastases patients, evaluating recurrence, mutational status, pathological characteristic (mucinous and tumor budding) and surgical resection margin. METHODS: This retrospective analysis was approved by the local Ethical Committee board, and radiological databases were used to select patients with colorectal liver metastases with pathological proof and MRI study in a pre-surgical setting after neoadjuvant chemotherapy. The cohort of patients included a training set (51 patients with 61 years of median age and 121 liver metastases) and an external validation set (30 patients with single lesion with 60 years of median age). For each segmented volume of interest on MRI by two expert radiologists, 851 radiomics features were extracted as median values using the PyRadiomics tool. Non-parametric Kruskal-Wallis test, intraclass correlation, receiver operating characteristic (ROC) analysis, linear regression modelling and pattern recognition methods (support vector machine (SVM), k-nearest neighbors (KNN), artificial neural network (NNET), and decision tree (DT)) were considered. RESULTS: The best predictor to discriminate expansive versus infiltrative tumor growth front was wavelet_LHH_glrlm_ShortRunLowGrayLevelEmphasis extracted on portal phase with accuracy of 82%, sensitivity of 84%, and specificity of 77%. The best predictor to discriminate tumor budding was wavelet_LLH_firstorder_10Percentile extracted on portal phase with accuracy of 92%, a sensitivity of 96%, and a specificity of 81%. The best predictor to differentiate the mucinous type of tumor was the wavelet_LLL_glcm_ClusterTendency extracted on portal phase with accuracy of 88%, a sensitivity of 38%, and a specificity of 100%. The best predictor to identify the recurrence was the wavelet_HLH_ngtdm_Complexity extracted on arterial phase with accuracy of 90%, a sensitivity of 71%, and a specificity of 95%. The best linear regression model was obtained in the identification of mucinous type considering the 13 textural significant metrics extracted by arterial phase (accuracy of 94%, sensitivity of 77% and a specificity of 99%). The best results were obtained in the identification of tumor budding with the eleven textural significant features extracted by arterial phase using a KNN (accuracy of 95%, sensitivity of 84%, and a specificity of 99%). CONCLUSIONS: Our results confirmed the capacity of radiomics to identify as biomarkers and several prognostic features that could affect the treatment choice in patients with liver metastases in order to obtain a more personalized approach.