1,721,113 research outputs found
An ensemble learning approach for brain cancer detection exploiting radiomic features
No full textBackground and Objective: The brain cancer is one of the most aggressive tumour: the 70% of the patients diagnosed with this malignant cancer will not survive. Early detection of brain tumours can be fundamental to increase survival rates. The brain cancers are classified into four different grades (i.e., I, II, III and IV) according to how normal or abnormal the brain cells look. The following work aims to recognize the different brain cancer grades by analysing brain magnetic resonance images. Methods: A method to identify the components of an ensemble learner is proposed. The ensemble learner is focused on the discrimination between different brain cancer grades using non invasive radiomic features. The considered radiomic features are belonging to five different groups: First Order, Shape, Gray Level Co-occurrence Matrix, Gray Level Run Length Matrix and Gray Level Size Zone Matrix. We evaluate the features effectiveness through hypothesis testing and through decision boundaries, performance analysis and calibration plots thus we select the best candidate classifiers for the ensemble learner. Results: We evaluate the proposed method with 111,205 brain magnetic resonances belonging to two freely available data-sets for research purposes. The results are encouraging: we obtain an accuracy of 99% for the benign grade I and the II, III and IV malignant brain cancer detection. Conclusion: The experimental results confirm that the ensemble learner designed with the proposed method outperforms the current state-of-the-art approaches in brain cancer grade detection starting from magnetic resonance images
Radiomics for gleason score detection through deep learning
No full textProstate cancer is classified into different stages, each stage is related to a different Gleason score. The labeling of a diagnosed prostate cancer is a task usually performed by radiologists. In this paper we propose a deep architecture, based on several convolutional layers, aimed to automatically assign the Gleason score to Magnetic Resonance Imaging (MRI) under analysis. We exploit a set of 71 radiomic features belonging to five categories: First Order, Shape, Gray Level Co-occurrence Matrix, Gray Level Run Length Matrix and Gray Level Size Zone Matrix. The radiomic features are gathered directly from segmented MRIs using two free-available dataset for research purpose obtained from different institutions. The results, obtained in terms of accuracy, are promising: they are ranging between 0.96 and 0.98 for Gleason score prediction
Explainable Deep Learning for Pulmonary Disease and Coronavirus COVID-19 Detection from X-rays
No full textBackground and Objective: Coronavirus disease (COVID-19) is an infectious disease caused by a new virus never identified before in humans. This virus causes respiratory disease (for instance, flu) with symptoms such as cough, fever and, in severe cases, pneumonia. The test to detect the presence of this virus in humans is performed on sputum or blood samples and the outcome is generally available within a few hours or, at most, days. Analysing biomedical imaging the patient shows signs of pneumonia. In this paper, with the aim of providing a fully automatic and faster diagnosis, we propose the adoption of deep learning for COVID-19 detection from X-rays. Method: In particular, we propose an approach composed by three phases: the first one to detect if in a chest X-ray there is the presence of a pneumonia. The second one to discern between COVID-19 and pneumonia. The last step is aimed to localise the areas in the X-ray symptomatic of the COVID-19 presence. Results and Conclusion: Experimental analysis on 6,523 chest X-rays belonging to different institutions demonstrated the effectiveness of the proposed approach, with an average time for COVID-19 detection of approximately 2.5 seconds and an average accuracy equal to 0.97
Formal methods for prostate cancer Gleason score and treatment prediction using radiomic biomarkers
No full textProstate cancer is a significant public health burden and a major cause of morbidity and mortality among men worldwide. Only in 2018 were reported 1.3 million of new diagnosed patients. Usually an invasive trans-perineal biopsy is the way to diagnose prostate cancer grade by prostate tissue removal. In this paper we propose a non invasive method to detect the prostate cancer grade (the so-called Gleason score) by computing radiomic biomarkers from magnetic resonance images. Furthermore, the proposed method predicts whether the cancer is suitable for the surgery treatment basing on the pathologist and surgeon suggestions. We represent patient magnetic resonances in terms of formal models and, through an algorithm designed by authors, we infer a set of properties aimed to predict the Gleason score and the treatment. By exploiting a formal verification environment, the properties are verified on two different real-world data-sets, the first one is composed of 36 patients, while the second one of 26, confirming the effectiveness of the proposed method
Three-dimensional endoanal ultrasound should be considered as first-line diagnostic tool in the preoperative work-up for perianal fistulas: The authors reply to the letter: Mathew RP, Patel V, Low G. Caution in using 3D-EAUS as the first-line diagnostic tool in the preoperative work up for perianal fistulas. Radiol Med 2020;125:155–156
No full textIn the preoperative work-up of patients with anorectal fistulas, 3D-EAUS may represent the first-line diagnostic tool, showing high diagnostic accuracy in the evaluation of internal openings, primary tracks and secondary extension. In the cases of fistulas classified as complex by 3D-EAUS, MRI may be indicated as adjunctive diagnostic imaging examination, to more accurately detect the fistulas’ secondary extensions, and so, to more carefully describe the fistulas’ complete anatomy
Intestinal Ischemia: US-CT findings correlations
Full text linkBackground: Intestinal ischemia is an abdominal emergency that accounts for approximately 2% of gastrointestinal illnesses. It represents a complex of diseases caused by impaired blood perfusion to the small and/ or large bowel including acute arterial mesenteric ischemia (AAMI), acute venous mesenteric ischemia (AVMI), non occlusive mesenteric ischemia (NOMI), ischemia/reperfusion injury (I/R), ischemic colitis (IC). In this study different study methods (US, CT) will be correlated in the detection of mesenteric ischemia imaging findings due to various etiologies. Methods: Basing on experience of our institutions, over 200 cases of mesenteric ischemia/infarction investigated with both US and CT were evaluated considering, in particular, the following findings: presence/absence of arterial/ venous obstruction, bowel wall thickness and enhancement, presence/absence of spastic reflex ileus, hypotonic reflex ileus or paralitic ileus, mural and/or portal/mesenteric pneumatosis, abdominal free fluid, parenchymal ischemia/infarction (liver, kidney, spleen). Results: To make an early diagnosis useful to ensure a correct therapeutic approach, it is very important to differentiate between occlusive (arterial,venous) and nonocclusive causes (NOMI). The typical findings of each forms of mesenteric ischemia are explained in the text. Conclusion: At present, the reference diagnostic modality for intestinal ischaemia is contrast-enhanced CT. However, there are some disadvantages associated with these techniques, such as radiation exposure, potential nephrotoxicity and the risk of an allergic reaction to the contrast agents. Thus, not all patients with suspected bowel ischaemia can be subjected to these examinations. Despite its limitations, US could constitutes a good imaging method as first examination in acute settings of suspected mesenteric ischemia. © 2013 Reginelli A et al; licensee BioMed Central Ltd
- …
