RESUMO
OBJECTIVES: To assess the value of contrast-enhanced (CE) diagnostic CT scans characterized through radiomics as predictors of recurrence for patients with stage II and III colorectal cancer in a two-center context. MATERIALS AND METHODS: This study included 193 patients diagnosed with stage II and III colorectal adenocarcinoma from 1 July 2008 to 15 March 2017 in two different French University Hospitals. To compensate for the variability in two-center data, a statistical harmonization method Bootstrapped ComBat (B-ComBat) was used. Models predicting disease-free survival (DFS) were built using 3 different machine learning (ML): (1) multivariate regression (MR) with 10-fold cross-validation after feature selection based on least absolute shrinkage and selection operator (LASSO), (2) random forest (RF), and (3) support vector machine (SVM), both with embedded feature selection. RESULTS: The performance for both balanced and 95% sensitivity models was systematically higher after our proposed B-ComBat harmonization compared to the use of the original untransformed data. The most clinically relevant performance was achieved by the multivariate regression model combining a clinical variable (postoperative chemotherapy) with two radiomics shape descriptors (compactness and least axis length) with a BAcc of 0.78 and an MCC of 0.6 associated with a required sensitivity of 95%. The resulting stratification in terms of DFS was significant (p = 0.00021), especially compared to the use of unharmonized original data (p = 0.17). CONCLUSIONS: Radiomics models derived from contrast-enhanced CT could be trained and validated in a two-center cohort with a good predictive performance of recurrence in stage II et III colorectal cancer patients. KEY POINTS: ⢠Adjuvant therapy decision in colorectal cancer can be a challenge in medical oncology. ⢠Radiomics models, derived from diagnostic CT, trained and validated in a two-center cohort, could predict recurrence in stage II and III colorectal cancer patients. ⢠Identifying patients with a low risk of recurrence, these models could facilitate treatment optimization and avoid unnecessary treatment.
Assuntos
Neoplasias Colorretais , Tomografia Computadorizada por Raios X , Neoplasias Colorretais/diagnóstico por imagem , Intervalo Livre de Doença , Humanos , Aprendizado de Máquina , Estudos Retrospectivos , Máquina de Vetores de SuporteRESUMO
BACKGROUND: Gastric pneumatosis (GP) is a rare radiologic finding with an unpredictable prognosis. The aim of this study was to identify mortality risk factors from patients presenting with GP on computed tomography (CT), and to develop a model which would allow us to predict which patients would benefit most from operative management. METHODS: Between 2010 and 2020, all CT-scan reports in four tertiary centers were searched for the following terms: "gastric pneumatosis," "intramural gastric air" or "emphysematous gastritis." The retrieved CT scans were reviewed by a senior surgeon and a senior radiologist. Relevant clinical and laboratory data for these patients were extracted from the institutions' medical records. RESULTS: Among 58 patients with GP, portal venous gas and bowel ischemia were present on CT scan in 52 (90%) and 17 patients (29%), respectively. The 30-day mortality rate was 31%. Univariate analysis identified the following variables as predictive of mortality at the time of the diagnosis of GP: abdominal guarding, hemodynamic instability, arterial lactate level >2 mmol/l, and the absence of gastric dilatation. Multivariable analysis identified the following variables as independent predictors of mortality: arterial lactate level (OR: 1.39, 95% CI: 1.07-1.79) and the absence of gastric dilatation (OR: 0.07, 95% CI: 0.01-0.79). None of the patients presenting with a baseline lactate rate<2 mmol/l died within 30 days following diagnosis, and no more than 17 patients out of 58 had bowel ischemia (29%). CONCLUSIONS: GP could be managed non-operatively, even in the presence of portal venous gas. However, patients with arterial lactate level>2 mmol/l, or the absence of gastric dilation should be surgically explored due to a non-negligible risk of mortality.
Assuntos
Dilatação Gástrica , Isquemia Mesentérica , Pneumatose Cistoide Intestinal , Humanos , Isquemia/diagnóstico por imagem , Isquemia/etiologia , Isquemia/cirurgia , Ácido Láctico , Pneumatose Cistoide Intestinal/diagnóstico por imagem , Pneumatose Cistoide Intestinal/terapia , Veia Porta/diagnóstico por imagem , Estudos RetrospectivosRESUMO
BACKGROUND: Eyelid basal cell carcinoma (BCC) is usually cured by surgery. However, for a minority of patients, extensive disease progression and recurrence contraindicate surgery or radiotherapy because of severe ocular morbidity. The hedgehog signaling pathway inhibitor vismodegib is becoming the key treatment for this specific form. OBJECTIVE: The aim of this review was to define the role of surgery after vismodegib treatment. MATERIALS AND METHODS: A literature search of the PubMed, Cochrane Library, ScienceDirect, and Embase databases was conducted for all articles published up to March 2021 to identify studies that examined treatment of BCC of the eyelid by vismodegib. RESULTS: Level 1 evidence was found for the use of vismodegib as neoadjuvant therapy in locally advanced eyelid BCC contraindicated to surgery and/or radiotherapy with a rather good tolerance of treatment. Level 3 evidence was found for the role of surgical excision of residual clinically suspicious lesions as for the indication of eyelid reconstruction after mapping or during residual tumoral resection if frozen sections or Mohs surgery were performed. CONCLUSION: Vismodegib is a well-tolerated treatment for advanced periorbital BCC. The hedgehog signaling pathway inhibitor vismodegib is a potential treatment option in patients with these challenging tumors.
Assuntos
Antineoplásicos , Carcinoma Basocelular , Neoplasias Cutâneas , Anilidas , Carcinoma Basocelular/tratamento farmacológico , Carcinoma Basocelular/cirurgia , Proteínas Hedgehog , Humanos , Piridinas , Neoplasias Cutâneas/tratamento farmacológico , Neoplasias Cutâneas/cirurgiaRESUMO
BACKGROUND: Progressive supranuclear palsy (PSP) is a neurodegenerative clinically heterogeneous disorder, formal diagnosis being based on postmortem histological brain examination. OBJECTIVE: We aimed to perform a precise in vivo staging of neurodegeneration in PSP using quantitative multimodal MRI. The ability of MRI biomarkers to differentiate PSP from PD was also evaluated. METHODS: Eleven PSP patients were compared to 26 age-matched healthy controls and 51 PD patients. Images were acquired at 3 Tesla (three-dimensional T1 -weighted, diffusion tensor, and neuromelanin-sensitive images) and 7 Tesla (three-dimensional-T2 * images). Regions of interest included the cortical areas, hippocampus, amygdala, basal ganglia, basal forebrain, brainstem nuclei, dentate nucleus, and cerebellum. Volumes, mean diffusivity, and fractional anisotropy were measured. In each region, a threshold value for group categorization was calculated, and four grades of change (0-3) were determined. RESULTS: PSP patients showed extensive volume decreases and diffusion changes in the midbrain, SN, STN, globus pallidus, basal forebrain, locus coeruleus, pedunculopontine nucleus, and dentate nucleus, in close agreement with the degrees of impairment in histological analyses. The predictive factors for the separation of PSP and healthy controls were, in descending order, the neuromelanin-based SN volume; midbrain fractional anisotropy; volumes of the midbrain, globus pallidus, and putamen; and fractional anisotropy in the locus coeruleus. The best predictors for separating PSP from PD were the neuromelanin-based volume in the SN, fractional anisotropy in the pons, volumes of the midbrain and globus pallidus, and fractional anisotropy in the basal forebrain. CONCLUSIONS: These results suggest that it is possible to evaluate brain neurodegeneration in PSP noninvasively, even in small brainstem nuclei, in close agreement with previously published histological data. © 2019 International Parkinson and Movement Disorder Society.
Assuntos
Encéfalo/patologia , Imagem de Difusão por Ressonância Magnética , Atrofia de Múltiplos Sistemas/patologia , Paralisia Supranuclear Progressiva/patologia , Idoso , Gânglios da Base/patologia , Imagem de Difusão por Ressonância Magnética/métodos , Feminino , Humanos , Masculino , Mesencéfalo/patologia , Pessoa de Meia-Idade , Doença de Parkinson/patologia , Transtornos Parkinsonianos/patologiaRESUMO
Augmented reality is a technology that allows a surgeon to see key hidden subsurface structures in an endoscopic video in real-time. This works by overlaying information obtained from preoperative imaging and fusing it in real-time with the endoscopic image. Magnetic resonance diffusion tensor imaging (DTI) and fiber tractography are known to provide additional information to that obtained from standard structural magnetic resonance imaging (MRI). Here, we report the first 2 cases of the use of real-time augmented reality during laparoscopic myomectomies with visualization of uterine muscle fibers after DTI tractography-MRI to help the surgeon decide the starting point incision. In the first case, a 31-year-old patient was undergoing laparoscopic surgery for a 6-cm FIGO type V myoma. In the second case, a 38-year-old patient was undergoing a laparoscopic myomectomy for a unique 6-cm FIGO type VI myoma. Signed consent forms were obtained from both patients, which included clauses of no modification of the surgery. Before surgery, MRI was performed. The external surface of the uterus, the uterine cavity, and the surface of the myomas were delimited on the basis of the findings of preoperative MRI. A fiber tracking algorithm was used to extrapolate the uterine muscle fibers' architecture. The aligned models were blended with each video frame to give the impression that the uterus is almost transparent, enabling the surgeon to localize the myomas and uterine cavity exactly. The uterine muscle fibers were also displayed, and their visualization helped us decide the starting incision point for the myomectomies. Then, myomectomies were performed using a classic laparoscopic technique. These case reports show that augmented reality and DTI fiber tracking in a uterus with myomas are possible, providing fiber direction and helping the surgeon visualize and decide the starting incision point for laparoscopic myomectomy. Respecting the fibers' orientation could improve the quality of the scar and decrease the architectural disorganization of the uterus.
Assuntos
Realidade Aumentada , Laparoscopia , Leiomioma , Mioma , Miomectomia Uterina , Neoplasias Uterinas , Adulto , Imagem de Tensor de Difusão , Feminino , Humanos , Laparoscopia/métodos , Leiomioma/diagnóstico por imagem , Leiomioma/patologia , Leiomioma/cirurgia , Mioma/cirurgia , Miomectomia Uterina/métodos , Neoplasias Uterinas/diagnóstico por imagem , Neoplasias Uterinas/patologia , Neoplasias Uterinas/cirurgiaRESUMO
Augmented reality (AR) is a surgical guidance technology that allows key hidden subsurface structures to be visualized by endoscopic imaging. We report here 2 cases of patients with adenomyoma selected for the AR technique. The adenomyomas were localized using AR during laparoscopy. Three-dimensional models of the uterus, uterine cavity, and adenomyoma were constructed before surgery from T2-weighted magnetic resonance imaging, allowing an intraoperative 3-dimensional shape of the uterus to be obtained. These models were automatically aligned and "fused" with the laparoscopic video in real time, giving the uterus a semitransparent appearance and allowing the surgeon in real time to both locate the position of the adenomyoma and uterine cavity and rapidly decide how best to access the adenomyoma. In conclusion, the use of our AR system designed for gynecologic surgery leads to improvements in laparoscopic adenomyomectomy and surgical safety.
Assuntos
Adenomioma/diagnóstico , Adenomioma/cirurgia , Realidade Aumentada , Procedimentos Cirúrgicos em Ginecologia/métodos , Cirurgia Assistida por Computador/métodos , Neoplasias Uterinas/diagnóstico , Neoplasias Uterinas/cirurgia , Adulto , Estudos de Viabilidade , Feminino , Humanos , Laparoscopia/métodos , Imageamento por Ressonância Magnética/métodosRESUMO
OBJECTIVES: Detection of liver metastases is crucial for guiding oncological management. Computed tomography through iterative reconstructions is widely used in this indication but has certain limitations. Deep learning image reconstructions (DLIR) use deep neural networks to achieve a significant noise reduction compared to iterative reconstructions. While reports have demonstrated improvements in image quality, their impact on liver metastases detection remains unclear. Our main objective was to determine whether DLIR affects the number of detected liver metastasis. Our secondary objective was to compare metastases conspicuity between the two reconstruction methods. METHODS: CT images of 121 patients with liver metastases were reconstructed using a 50% adaptive statistical iterative reconstruction (50%-ASiR-V), and three levels of DLIR (DLIR-low, DLIR-medium, and DLIR-high). For each reconstruction, two double-blinded radiologists counted up to a maximum of ten metastases. Visibility and contour definitions were also assessed. Comparisons between methods for continuous parameters were performed using mixed models. RESULTS: A higher number of metastases was detected by one reader with DLIR-high: 7 (2-10) (median (Q1-Q3); total 733) versus 5 (2-10), respectively for DLIR-medium, DLIR-low, and ASiR-V (p < 0.001). Ten patents were detected with more metastases with DLIR-high simultaneously by both readers and a third reader for confirmation. Metastases visibility and contour definition were better with DLIR than ASiR-V. CONCLUSION: DLIR-high enhanced the detection and visibility of liver metastases compared to ASiR-V, and also increased the number of liver metastases detected. CRITICAL RELEVANCE STATEMENT: Deep learning-based reconstruction at high strength allowed an increase in liver metastases detection compared to hybrid iterative reconstruction and can be used in clinical oncology imaging to help overcome the limitations of CT. KEY POINTS: Detection of liver metastases is crucial but limited with standard CT reconstructions. More liver metastases were detected with deep-learning CT reconstruction compared to iterative reconstruction. Deep learning reconstructions are suitable for hepatic metastases staging and follow-up.
RESUMO
Richter transformation (RT) is an aggressive lymphoma occurring in patients with chronic lymphocytic leukaemia. Here we investigated the anti-CD3/anti-CD19 T-cell-engager blinatumomab after R-CHOP (i.e. rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone) in patients with untreated RT of diffuse large B-cell lymphoma histology (NCT03931642). In this multicentre phase 2 study, patients without complete response (CR) after two cycles of R-CHOP were eligible to receive an 8-week blinatumomab induction via continuous vein infusion with stepwise dosing until 112 µg/day. The primary endpoint was the CR rate after blinatumomab induction and secondary endpoint included safety, response duration, progression-free and overall survival. Thirty-nine patients started the first cycle of R-CHOP, 25 of whom received blinatumomab. After blinatumomab induction, five (20%) patients achieved CR, four (16%) achieved partial response, and six (24%) were stable. Considering the entire strategy, the overall response rate in the full-analysis-set was 46% (n = 18), with CR in 14 (36%) patients. The most common treatment-emergent adverse events of all grades in the blinatumomab-safety-set included fever (36%), anaemia (24%), and lymphopaenia (24%). Cytokine release syndrome (grade 1/2) was observed in 16% and neurotoxicity in 20% of patients. Blinatumomab demonstrated encouraging anti-tumour activity (the trial met its primary endpoint) and acceptable toxicity in patients with RT.
Assuntos
Anticorpos Biespecíficos , Protocolos de Quimioterapia Combinada Antineoplásica , Ciclofosfamida , Doxorrubicina , Linfoma Difuso de Grandes Células B , Prednisona , Rituximab , Vincristina , Humanos , Masculino , Feminino , Anticorpos Biespecíficos/administração & dosagem , Anticorpos Biespecíficos/efeitos adversos , Anticorpos Biespecíficos/uso terapêutico , Protocolos de Quimioterapia Combinada Antineoplásica/uso terapêutico , Protocolos de Quimioterapia Combinada Antineoplásica/efeitos adversos , Pessoa de Meia-Idade , Ciclofosfamida/uso terapêutico , Ciclofosfamida/administração & dosagem , Ciclofosfamida/efeitos adversos , Rituximab/administração & dosagem , Rituximab/uso terapêutico , Rituximab/efeitos adversos , Doxorrubicina/uso terapêutico , Doxorrubicina/administração & dosagem , Doxorrubicina/efeitos adversos , Idoso , Prednisona/uso terapêutico , Prednisona/administração & dosagem , Prednisona/efeitos adversos , Vincristina/uso terapêutico , Vincristina/efeitos adversos , Vincristina/administração & dosagem , Linfoma Difuso de Grandes Células B/tratamento farmacológico , Adulto , Idoso de 80 Anos ou mais , Leucemia Linfocítica Crônica de Células B/tratamento farmacológico , Resultado do TratamentoRESUMO
BACKGROUND: Non-alcoholic fatty liver disease (NAFLD) is becoming a major health problem, resulting in hepatic, metabolic and cardio-vascular morbidity. AIM: To evaluate new ultrasonographic tools to detect and measure hepatic steatosis. METHODS: We prospectively included 105 patients referred to our liver unit for NAFLD suspicion or follow-up. They underwent ultrasonographic measurement of liver sound speed estimation (SSE) and attenuation coefficient (AC) using Aixplorer MACH 30 (Supersonic Imagine, France), continuous controlled attenuation parameter (cCAP) using Fibroscan (Echosens, France) and standard liver ultrasound with hepato-renal index (HRI) calculation. Hepatic steatosis was then classified according to magnetic resonance imaging proton density fat fraction (PDFF). Receiver operating curve (ROC) analysis was performed to evaluate the diagnostic performance in the diagnosis of steatosis. RESULTS: Most patients were overweight or obese (90%) and had metabolic syndrome (70%). One third suffered from diabetes. Steatosis was identified in 85 patients (81%) according to PDFF. Twenty-one patients (20%) had advanced liver disease. SSE, AC, cCAP and HRI correlated with PDFF, with respective Spearman correlation coefficient of -0.39, 0.42, 0.54 and 0.59 (P < 0.01). Area under the receiver operating characteristic curve (AUROC) for detection of steatosis with HRI was 0.91 (0.83-0.99), with the best cut-off value being 1.3 (Se = 83%, Sp = 98%). The optimal cCAP threshold of 275 dB/m, corresponding to the recent EASL-suggested threshold, had a sensitivity of 72% and a specificity of 80%. Corresponding AUROC was 0.79 (0.66-0.92). The diagnostic accuracy of cCAP was more reliable when standard deviation was < 15 dB/m with an AUC of 0.91 (0.83-0.98). An AC threshold of 0.42 dB/cm/MHz had an AUROC was 0.82 (0.70-0.93). SSE performed moderately with an AUROC of 0.73 (0.62-0.84). CONCLUSION: Among all ultrasonographic tools evaluated in this study, including new-generation tools such as cCAP and SSE, HRI had the best performance. It is also the simplest and most available method as most ultrasound scans are equipped with this module.
Assuntos
Hepatopatia Gordurosa não Alcoólica , Humanos , Hepatopatia Gordurosa não Alcoólica/diagnóstico por imagem , Estudos Prospectivos , Ultrassonografia , Imageamento por Ressonância MagnéticaRESUMO
PURPOSE: Immunotherapy has dramatically improved the prognosis of patients with metastatic melanoma (MM). Yet, there is a lack of biomarkers to predict whether a patient will benefit from immunotherapy. Our aim was to create radiomics models on pretreatment computed tomography (CT) to predict overall survival (OS) and treatment response in patients with MM treated with anti-PD-1 immunotherapy. METHODS: We performed a monocentric retrospective analysis of 503 metastatic lesions in 71 patients with 46 radiomics features extracted following lesion segmentation. Predictive accuracies for OS < 1 year versus > 1 year and treatment response versus no response was compared for five feature selection methods (sequential forward selection, recursive, Boruta, relief, random forest) and four classifiers (support vector machine (SVM), random forest, K-nearest neighbor, logistic regression (LR)) used with or without SMOTE data augmentation. A fivefold cross-validation was performed at the patient level, with a tumour-based classification. RESULTS: The highest accuracy level for OS predictions was obtained with 3D lesions (0.91) without clinical data integration when combining Boruta feature selection and the LR classifier, The highest accuracy for treatment response prediction was obtained with 3D lesions (0.88) without clinical data integration when combining Boruta feature selection, the LR classifier and SMOTE data augmentation. The accuracy was significantly higher concerning OS prediction with 3D segmentation (0.91 vs 0.86) while clinical data integration led to improved accuracy notably in 2D lesions (0.76 vs 0.87) regarding treatment response prediction. Skewness was the only feature found to be an independent predictor of OS (HR (CI 95%) 1.34, p-value 0.001). CONCLUSION: This is the first study to investigate CT texture parameter selection and classification methods for predicting MM prognosis with treatment by immunotherapy. Combining pretreatment CT radiomics features from a single tumor with data selection and classifiers may accurately predict OS and treatment response in MM treated with anti-PD-1.
Assuntos
Melanoma , Humanos , Imunoterapia , Melanoma/diagnóstico por imagem , Melanoma/terapia , Prognóstico , Estudos Retrospectivos , Tomografia Computadorizada por Raios X/métodosRESUMO
BACKGROUND: Endoscopic mucosal healing is the current therapeutic target in Crohn's disease. However, transmural healing could lead to better outcomes. AIMS: To assess whether transmural healing or magnetic resonance imaging (MRI) healing are better therapeutic targets than endoscopic mucosal healing to predict long-term improved outcome in Crohn's disease METHODS: From our MRI database, we retrospectively identified all Crohn's disease patients who had MRI and colonoscopy within a 3-month interval (median interval = 17.5 days). Four groups were considered: endoscopic mucosal healing (no ulceration or aphthoid erosion), MRI healing (no MRI signs of inflammation and no complication), transmural healing (combination of endoscopic and MRI healing) or no healing. Outcomes were time to surgery, bowel damage progression, hospitalisation, major outcomes (one of the three previous endpoints) and Crohn's disease-related drug discontinuation. Results were expressed in multivariable analyses adjusted on potential confounders (hazard ratio (HR) [95% confidence interval]). RESULTS: Among 154 patients with Crohn's disease, 51.9% (80/154), 10.4% (16/154), 19.5% (30/154) and 18.2% (28/154) achieved no healing, endoscopic mucosal healing, MRI healing and transmural healing, respectively. Transmural healing (HR = 0.05 [0.00-0.40], P = 0.002) and MRI healing (HR = 0.09 [0.00-0.47], P = 0.005) were associated with lower risk of bowel damage progression than endoscopic mucosal healing. In addition, achieving transmural healing or MRI healing reduced the risk of experiencing major outcomes compared to endoscopic mucosal healing (HR = 0.28 [0.00-0.74], P = 0.01). Patients with transmural healing also had a decreased risk of relapse-related drug discontinuation (HR = 0.35 [0.13-0.95], P = 0.039) compared to those with endoscopic mucosal healing. CONCLUSION: Transmural healing and MRI healing are associated with lower risk of bowel damage progression than endoscopic mucosal healing and could be considered as better therapeutic targets in Crohn's disease.
Assuntos
Doença de Crohn , Doença de Crohn/diagnóstico por imagem , Humanos , Mucosa Intestinal/diagnóstico por imagem , Imageamento por Ressonância Magnética , Estudos Prospectivos , Estudos RetrospectivosRESUMO
PURPOSE: We present and evaluate a new automated method based on support vector machine (SVM) classification of whole-brain anatomical magnetic resonance imaging to discriminate between patients with Alzheimer's disease (AD) and elderly control subjects. MATERIALS AND METHODS: We studied 16 patients with AD [mean age +/- standard deviation (SD) = 74.1 +/- 5.2 years, mini-mental score examination (MMSE) = 23.1 +/- 2.9] and 22 elderly controls (72.3 +/- 5.0 years, MMSE = 28.5 +/- 1.3). Three-dimensional T1-weighted MR images of each subject were automatically parcellated into regions of interest (ROIs). Based upon the characteristics of gray matter extracted from each ROI, we used an SVM algorithm to classify the subjects and statistical procedures based on bootstrap resampling to ensure the robustness of the results. RESULTS: We obtained 94.5% mean correct classification for AD and control subjects (mean specificity, 96.6%; mean sensitivity, 91.5%). CONCLUSIONS: Our method has the potential in distinguishing patients with AD from elderly controls and therefore may help in the early diagnosis of AD.
Assuntos
Algoritmos , Doença de Alzheimer/patologia , Inteligência Artificial , Encéfalo/patologia , Interpretação de Imagem Assistida por Computador/métodos , Imageamento por Ressonância Magnética/métodos , Reconhecimento Automatizado de Padrão/métodos , Idoso , Feminino , Humanos , Aumento da Imagem/métodos , Masculino , Reprodutibilidade dos Testes , Sensibilidade e EspecificidadeRESUMO
Developing methods to segment the liver in medical images, study and analyze it remains a significant challenge. The shape of the liver can vary considerably from one patient to another, and adjacent organs are visualized in medical images with similar intensities, making the boundaries of the liver ambiguous. Consequently, automatic or semi-automatic segmentation of liver is a difficult task. Moreover, scanning systems and magnetic resonance imaging have different settings and parameters. Thus the images obtained differ from one machine to another. In this article, we propose an automatic model-based segmentation that allows building a faithful 3-D representation of the liver, with a mean Dice value equal to 90.3% on CT and MRI datasets. We compare our algorithm with a semi-automatic method and with other approaches according to the state of the art. Our method works with different data sources, we use a large quantity of CT and MRI images from machines in various hospitals and multiple DICOM images available from public challenges. Finally, for evaluation of liver segmentation approaches in state of the art, robustness is not adequacy addressed with a precise definition. Another originality of this article is the introduction of a novel measure of robustness, which takes into account the liver variability at different scales.
Assuntos
Algoritmos , Processamento de Imagem Assistida por Computador/métodos , Hepatopatias/diagnóstico por imagem , Fígado/diagnóstico por imagem , Imageamento por Ressonância Magnética , Tomografia Computadorizada por Raios X , Conjuntos de Dados como Assunto , Feminino , Humanos , Imageamento Tridimensional , MasculinoRESUMO
BACKGROUND: Proper segmentation of the liver from medical images is critical for computer-assisted diagnosis, therapy and surgical planning. Knowledge of its vascular structure allows division of the liver into eight functionally independent segments, each with its own vascular inflow, known as the Couinaud scheme. Couinaud's description is the most widely used classification, since it is well-suited for surgery and accurate for the localization of lesions. However, automatic segmentation of the liver and its vascular structure to construct the Couinaud scheme remains a challenging task. METHODS: We present a complete framework to obtain Couinaud's classification in three main steps; first, we propose a model-based liver segmentation, then a vascular segmentation based on a skeleton process, and finally, the construction of the eight independent liver segments. Our algorithms are automatic and allow 3D visualizations. RESULTS: We validate these algorithms on various databases with different imaging modalities (Magnetic Resonance Imaging (MRI) and Computed Tomography (CT)). Experimental results are presented on diseased livers, which pose complex challenges because both the overall organ shape and the vessels can be severely deformed. A mean DICE score of 0.915 is obtained for the liver segmentation, and an average accuracy of 0.98 for the vascular network. Finally, we present an evaluation of our method for performing the Couinaud segmentation thanks to medical reports with promising results. CONCLUSIONS: We were able to automatically reconstruct 3-D volumes of the liver and its vessels on MRI and CT scans. Our goal is to develop an improved method to help radiologists with tumor localization.
Assuntos
Algoritmos , Imageamento Tridimensional , Fígado , Angiografia por Ressonância Magnética , Tomografia Computadorizada por Raios X , Humanos , Fígado/irrigação sanguínea , Fígado/diagnóstico por imagemRESUMO
PURPOSE: Given the growing prevalence of obesity and metabolic syndrome, the management of hepatic steatosis, especially its quantification, is a major issue. We assessed the quantification of liver steatosis using four different MR methods, in order to determine the one that is best correlated with the reference method which consists of histological measurement by liver biopsy. METHOD: Seventy-one successive patients requiring liver biopsy for acute or chronic liver disease were enrolled prospectively between March 2017 and March 2018, 11 were excluded and 60 were reported. Liver MR (1.5â¯T) was organised in order to be performed the same day, using four different steatosis quantification techniques (3-echo MRI, 6-echo MRI, 11-echo MRI and MR Spectroscopy). Quantitative histological and imaging data were compared. In a secondary analysis, we studied the possible influence of alcohol drinking, hepatic iron overload, and the presence of liver fibrosis. RESULTS: All four MR techniques were found to have excellent correlations with the histological measurements: 3-echo MRI (râ¯=â¯0.852, pâ¯<â¯0.001), 6-echo MRI (râ¯=â¯0.819, pâ¯<â¯0.001), 11-echo MRI (râ¯=â¯0.818, pâ¯<â¯0.001) and MR Spectroscopy (râ¯=â¯0,812, pâ¯<â¯0,001). Interestingly, we also found that the presence of alcohol consumption, iron overload and fibrosis did not interfere with measurements, whichever technique was used. CONCLUSION: In the evaluation of hepatic steatosis, our study showed very good correlations of all four MR techniques with the histological standard. There was no confounding factor in a representative group of patients with associated liver conditions such as alcohol consumption, fibrosis and iron overload, for each technique. All four MR techniques may be used in daily practice.
Assuntos
Fígado Gorduroso/diagnóstico por imagem , Fígado Gorduroso/patologia , Imageamento por Ressonância Magnética/métodos , Espectroscopia de Ressonância Magnética/métodos , Hepatopatia Gordurosa não Alcoólica/diagnóstico por imagem , Hepatopatia Gordurosa não Alcoólica/patologia , Adulto , Idoso , Idoso de 80 Anos ou mais , Biópsia , Feminino , Humanos , Fígado/diagnóstico por imagem , Fígado/patologia , Masculino , Pessoa de Meia-Idade , Reprodutibilidade dos TestesRESUMO
PURPOSE: To prospectively evaluate the accuracy of automated hippocampal volumetry to help distinguish between patients with Alzheimer disease (AD), patients with mild cognitive impairment (MCI), and elderly controls, by using established criteria for patients with AD and MCI as the reference standard. MATERIALS AND METHODS: The regional ethics committee approved the study and written informed consent was obtained from all participants. The study included 25 patients with AD (11 men, 14 women; mean age +/- standard deviation [SD], 73 years +/- 6; Mini-Mental State Examination (MMSE) score, 24.4 +/- 2.7), 24 patients with amnestic MCI (10 men, 14 women; mean age +/- SD, 74 years +/- 8; MMSE score, 27.2 +/- 1.4) and 25 elderly healthy controls (13 men, 12 women; mean age +/- SD, 64 years +/- 8). For each participant, the hippocampi were automatically segmented on three-dimensional T1-weighted magnetic resonance (MR) images with high spatial resolution. Segmentation was performed by using recently developed software that allows fast segmentation with minimal user input. Group differences in hippocampal volume were assessed by using Student t tests. To obtain robust estimates of P values, the correct classification rate, sensitivity, and specificity, bootstrap methods were used. RESULTS: Significant hippocampal volume reductions were detected in all groups of patients (-32% in AD patients vs controls, P < .001; -19% in MCI patients vs controls, P < .001; and -15% in AD patients vs MCI patients, P < .01). Individual classification on the basis of hippocampal volume resulted in 84% correct classification (sensitivity, 84%; specificity, 84%) between AD patients and controls and 73% correct classification (sensitivity, 75%; specificity, 70%) between MCI patients and controls. CONCLUSION: This automated method can serve as an alternative to manual tracing and may thus prove useful in assisting with the diagnosis of AD.
Assuntos
Envelhecimento/patologia , Doença de Alzheimer/diagnóstico , Transtornos Cognitivos/diagnóstico , Hipocampo/patologia , Interpretação de Imagem Assistida por Computador/métodos , Imageamento por Ressonância Magnética/métodos , Reconhecimento Automatizado de Padrão/métodos , Idoso , Algoritmos , Inteligência Artificial , Diagnóstico Diferencial , Feminino , Humanos , Aumento da Imagem/métodos , Masculino , Pessoa de Meia-Idade , Reprodutibilidade dos Testes , Sensibilidade e EspecificidadeRESUMO
AIM: To assess magnetic resonance imaging (MRI) and faecal calprotectin to detect endoscopic postoperative recurrence in patients with Crohn's disease (CD). METHODS: From two tertiary centers, all patients with CD who underwent ileocolonic resection were consecutively and prospectively included. All the patients underwent MRI and endoscopy within the first year after surgery or after the restoration of intestinal continuity [median = 6 mo (5.0-9.3)]. The stools were collected the day before the colonoscopy to evaluate faecal calprotectin level. Endoscopic postoperative recurrence (POR) was defined as Rutgeerts' index ≥ i2b. The MRI was analyzed independently by two radiologists blinded from clinical data. RESULTS: Apparent diffusion coefficient (ADC) was lower in patients with endoscopic POR compared to those with no recurrence (2.03 ± 0.32 vs 2.27 ± 0.38 × 10-3 mm²/s, P = 0.032). Clermont score (10.4 ± 5.8 vs 7.4 ± 4.5, P = 0.038) and relative contrast enhancement (RCE) (129.4% ± 62.8% vs 76.4% ± 32.6%, P = 0.007) were significantly associated with endoscopic POR contrary to the magnetic resonance index of activity (MaRIA) (7.3 ± 4.5 vs 4.8 ± 3.7; P = 0.15) and MR scoring system (P = 0.056). ADC < 2.35 × 10-3 mm²/s [sensitivity = 0.85, specificity = 0.65, positive predictive value (PPV) = 0.85, negative predictive value (NPV) = 0.65] and RCE > 100% (sensitivity = 0.75, specificity = 0.81, PPV = 0.75, NPV = 0.81) were the best cut-off values to identify endoscopic POR. Clermont score > 6.4 (sensitivity = 0.61, specificity = 0.82, PPV = 0.73, NPV = 0.74), MaRIA > 3.76 (sensitivity = 0.61, specificity = 0.82, PPV = 0.73, NPV = 0.74) and a MR scoring system ≥ MR1 (sensitivity = 0.54, specificity = 0.82, PPV = 0.70, and NPV = 0.70) demonstrated interesting performances to detect endoscopic POR. Faecal calprotectin values were significantly higher in patients with endoscopic POR (114 ± 54.5 µg/g vs 354.8 ± 432.5 µg/g; P = 0.0075). Faecal calprotectin > 100 µg/g demonstrated high performances to detect endoscopic POR (sensitivity = 0.67, specificity = 0.93, PPV = 0.89 and NPV = 0.77). CONCLUSION: Faecal calprotectin and MRI are two reliable tools to detect endoscopic POR in patients with CD.