Development of a Secure Web-Based Medical Imaging Analysis Platform: The AWESOMME Project.

Precision medicine research benefits from machine learning in the creation of robust models adapted to the processing of patient data. This applies both to pathology identification in images, i.e., annotation or segmentation, and to computer-aided diagnostic for classification or prediction. It comes with the strong need to exploit and visualize large volumes of images and associated medical data. The work carried out in this paper follows on from a main case study piloted in a cancer center. It proposes an analysis pipeline for patients with osteosarcoma through segmentation, feature extraction and application of a deep learning model to predict response to treatment. The main aim of the AWESOMME project is to leverage this work and implement the pipeline on an easy-to-access, secure web platform. The proposed WEB application is based on a three-component architecture: a data server, a heavy computation and authentication server and a medical imaging web-framework with a user interface. These existing components have been enhanced to meet the needs of security and traceability for the continuous production of expert data. It innovates by covering all steps of medical imaging processing (visualization and segmentation, feature extraction and aided diagnostic) and enables the test and use of machine learning models. The infrastructure is operational, deployed in internal production and is currently being installed in the hospital environment. The extension of the case study and user feedback enabled us to fine-tune functionalities and proved that AWESOMME is a modular solution capable to analyze medical data and share research algorithms with in-house clinicians.

Towards tumour hypoxia imaging: Incorporating relative oxygen extraction fraction mapping of prostate with multi-parametric quantitative MRI on a 1.5T MR-linac.

Hypoxia plays a central role in tumour radioresistance. Reliable tumour hypoxia imaging would allow the monitoring of tumour response and a more personalized adaptation of radiotherapy planning. Here, we showed a proof of concept of the feasibility and repeatability of relative oxygen extraction fraction (rOEF) mapping of prostate using multi-parametric quantitative MRI (qMRI) achieved for the first time on a 1.5T MR-linac. T2, T2* relaxation times maps, and intra-voxel incoherent motion (IVIM) parametric maps mapping were computed on a 29 years old healthy volunteer. R2' and rOEF maps were calculated based on a multi-parametric model. Long-term repeatability and repeatability coefficient (RC) were determined for each parameter according to QIBA recommendations. Mean values for the entire healthy prostate were 0.99 ± 0.14 × 10-3 mm/s2, 81 ± 2.1 × 10-3 mm/s2, 21.6 ± 3.6%, 92.7 ± 19.7 ms and 62.4 ± 17.3 ms for Dslow, Dfast, f, T2 and T2*, respectively. R2' and rOEF in the prostate were 6.1 ± 3.4 s-1 and 18.2 ± 10.1% respectively. The RC of rOEF was 4.43%. Long-term repeatability of quantitative parameters based on a test-retest ranged from 2 to 18%. qMRI parameters are measurable and repeatable on 1.5T MR LINAC. From T2, T2* and IVIM parameters maps, we were able to obtain a rOEF mapping of the prostate. These results are the first step to a non-invasive imaging of tumour hypoxia during radiotherapy leading to a biological image-guided adaptive radiotherapy.

Intravoxel incoherent motion magnetic resonance imaging to assess early tumor response to radiation therapy: Review and future directions.

Early prediction of radiation response by imaging is a dynamic field of research and it can be obtained using a variety of noninvasive magnetic resonance imaging methods. Recently, intravoxel incoherent motion (IVIM) has gained interest in cancer imaging. IVIM carries both diffusion and perfusion information, making it a promising tool to assess tumor response. Here, we briefly introduced the basics of IVIM, reviewed existing studies of IVIM in various type of tumors during radiotherapy in order to show whether IVIM is a useful technique for an early assessment of radiation response. 31/40 studies reported an increase of IVIM parameters during radiotherapy compared to baseline. In 27 studies, this increase was higher in patients with good response to radiotherapy. Future directions including implementation of IVIM on MR-Linac and its limitation are discussed. Obtaining new radiologic biomarkers of radiotherapy response could open the way for a more personalized, biology-guided radiation therapy.

Characterization of Breast Tumors from MR Images Using Radiomics and Machine Learning Approaches.

Determining histological subtypes, such as invasive ductal and invasive lobular carcinomas (IDCs and ILCs) and immunohistochemical markers, such as estrogen response (ER), progesterone response (PR), and the HER2 protein status is important in planning breast cancer treatment. MRI-based radiomic analysis is emerging as a non-invasive substitute for biopsy to determine these signatures. We explore the effectiveness of radiomics-based and CNN (convolutional neural network)-based classification models to this end. T1-weighted dynamic contrast-enhanced, contrast-subtracted T1, and T2-weighted MR images of 429 breast cancer tumors from 323 patients are used. Various combinations of input data and classification schemes are applied for ER+ vs. ER-, PR+ vs. PR-, HER2+ vs. HER2-, and IDC vs. ILC classification tasks. The best results were obtained for the ER+ vs. ER- and IDC vs. ILC classification tasks, with their respective AUCs reaching 0.78 and 0.73 on test data. The results with multi-contrast input data were generally better than the mono-contrast alone. The radiomics and CNN-based approaches generally exhibited comparable results. ER and IDC/ILC classification results were promising. PR and HER2 classifications need further investigation through a larger dataset. Better results by using multi-contrast data might indicate that multi-parametric quantitative MRI could be used to achieve more reliable classifiers.

Radiomics combined with transcriptomics to predict response to immunotherapy from patients treated with PD-1/PD-L1 inhibitors for advanced NSCLC.

Introduction: In this study, we aim to build radiomics and multiomics models based on transcriptomics and radiomics to predict the response from patients treated with the PD-L1 inhibitor. Materials and methods: One hundred and ninety-five patients treated with PD-1/PD-L1 inhibitors were included. For all patients, 342 radiomic features were extracted from pretreatment computed tomography scans. The training set was built with 110 patients treated at the Léon Bérard Cancer Center. An independent validation cohort was built with the 85 patients treated in Dijon. The two sets were dichotomized into two classes, patients with disease control and those considered non-responders, in order to predict the disease control at 3 months. Various models were trained with different feature selection methods, and different classifiers were evaluated to build the models. In a second exploratory step, we used transcriptomics to enrich the database and develop a multiomic signature of response to immunotherapy in a 54-patient subgroup. Finally, we considered the HOT/COLD status. We first trained a radiomic model to predict the HOT/COLD status and then prototyped a hybrid model integrating radiomics and the HOT/COLD status to predict the response to immunotherapy. Results: Radiomic signature for 3 months' progression-free survival (PFS) classification: The most predictive model had an area under the receiver operating characteristic curve (AUROC) of 0.94 on the training set and 0.65 on the external validation set. This model was obtained with the t-test selection method and with a support vector machine (SVM) classifier. Multiomic signature for PFS classification: The most predictive model had an AUROC of 0.95 on the training set and 0.99 on the validation set. Radiomic model to predict the HOT/COLD status: the most predictive model had an AUROC of 0.93 on the training set and 0.86 on the validation set. HOT/COLD radiomic hybrid model for PFS classification: the most predictive model had an AUROC of 0.93 on the training set and 0.90 on the validation set. Conclusion: In conclusion, radiomics could be used to predict the response to immunotherapy in non-small-cell lung cancer patients. The use of transcriptomics or the HOT/COLD status, together with radiomics, may improve the working of the prediction models.

Prediction of Histologic Neoadjuvant Chemotherapy Response in Osteosarcoma Using Pretherapeutic MRI Radiomics.

Histologic response to chemotherapy for osteosarcoma is one of the most important prognostic factors for survival, but assessment occurs after surgery. Although tumor imaging is used for surgical planning and follow-up, it lacks predictive value. Therefore, a radiomics model was developed to predict the response to neoadjuvant chemotherapy based on pretreatment T1-weighted contrast-enhanced MRI. A total of 176 patients (median age, 20 years [range, 5-71 years]; 107 male patients) with osteosarcoma treated with neoadjuvant chemotherapy and surgery between January 2007 and December 2018 in three different centers in France (Centre Léon Bérard in Lyon, Centre Hospitalier Universitaire de Nantes in Nantes, and Hôpital Cochin in Paris) were retrospectively analyzed. Various models were trained from different configurations of the data sets. Two different methods of feature selection were tested with and without ComBat harmonization (ReliefF and t test) to select the most relevant features, and two different classifiers were used to build the models (an artificial neural network and a support vector machine). Sixteen radiomics models were built using the different combinations of feature selection and classifier applied on the various data sets. The most predictive model had an area under the receiver operating characteristic curve of 0.95, a sensitivity of 91%, and a specificity 92% in the training set; respective values in the validation set were 0.97, 91%, and 92%. In conclusion, MRI-based radiomics may be useful to stratify patients receiving neoadjuvant chemotherapy for osteosarcomas. Keywords: MRI, Skeletal-Axial, Oncology, Radiomics, Osteosarcoma, Pediatrics Supplemental material is available for this article. © RSNA, 2022.

Prediction of lipomatous soft tissue malignancy on MRI: comparison between machine learning applied to radiomics and deep learning.

OBJECTIVES: Malignancy of lipomatous soft-tissue tumours diagnosis is suspected on magnetic resonance imaging (MRI) and requires a biopsy. The aim of this study is to compare the performances of MRI radiomic machine learning (ML) analysis with deep learning (DL) to predict malignancy in patients with lipomas oratypical lipomatous tumours. METHODS: Cohort include 145 patients affected by lipomatous soft tissue tumours with histology and fat-suppressed gadolinium contrast-enhanced T1-weighted MRI pulse sequence. Images were collected between 2010 and 2019 over 78 centres with non-uniform protocols (three different magnetic field strengths (1.0, 1.5 and 3.0 T) on 16 MR systems commercialised by four vendors (General Electric, Siemens, Philips, Toshiba)). Two approaches have been compared: (i) ML from radiomic features with and without batch correction; and (ii) DL from images. Performances were assessed using 10 cross-validation folds from a test set and next in external validation data. RESULTS: The best DL model was obtained using ResNet50 (resulting into an area under the curve (AUC) of 0.87 ± 0.11 (95% CI 0.65-1). For ML/radiomics, performances reached AUCs equal to 0.83 ± 0.12 (95% CI 0.59-1) and 0.99 ± 0.02 (95% CI 0.95-1) on test cohort using gradient boosting without and with batch effect correction, respectively. On the external cohort, the AUC of the gradient boosting model was equal to 0.80 and for an optimised decision threshold sensitivity and specificity were equal to 100% and 32% respectively. CONCLUSIONS: In this context of limited observations, batch-effect corrected ML/radiomics approaches outperformed DL-based models.

Multicontrast MRI-based radiomics for the prediction of pathological complete response to neoadjuvant chemotherapy in patients with early triple negative breast cancer.

INTRODUCTION: To assess pre-therapeutic MRI-based radiomic analysis to predict the pathological complete response to neoadjuvant chemotherapy (NAC) in women with early triple negative breast cancer (TN). MATERIALS AND METHODS: This monocentric retrospective study included 75 TN female patients with MRI (T1-weighted, T2-weighted, diffusion-weighted and dynamic contrast enhancement images) performed before NAC. For each patient, the tumor(s) and the parenchyma were independently segmented and analyzed with radiomic analysis to extract shape, size, and texture features. Several sets of features were realized based on the 4 different sequence images. Performances of 4 classifiers (random forest, multilayer perceptron, support vector machine (SVM) with linear or quadratic kernel) were compared based on pathological complete response (defined on the excised tissues), on 100 draws with 75% as training set and 25% as test. RESULTS: The combination of features extracted from different MR images improved the classifier performance (more precisely, the features from T1W, T2W and DWI). The SVM with quadratic kernel showed the best performance with a mean AUC of 0.83, a sensitivity of 0.85 and a specificity of 0.75 in the test set. CONCLUSION: MRI-based radiomics may be relevant to predict NAC response in TN cancer. Our results promote the use of multi-contrast MRI sources for radiomics, providing enrich source of information to enhance model generalization.

MRI-based radiomics to predict lipomatous soft tissue tumors malignancy: a pilot study.

OBJECTIVES: To develop and validate a MRI-based radiomic method to predict malignancies in lipomatous soft tissue tumors. METHODS: This retrospective study searched in the database of our pathology department, data from patients with lipomatous soft tissue tumors, with histology and gadolinium-contrast enhanced T1w MR images, obtained from 56 centers with non-uniform protocols. For each tumor, 87 radiomic features were extracted by two independent observers to evaluate the inter-observer reproducibility. A reduction of learning base dimension was performed from reproducibility and relevancy criteria. A model was subsequently prototyped using a linear support vector machine to predict malignant lesions. RESULTS: Eighty-one subjects with lipomatous soft tissue tumors including 40 lipomas and 41 atypical lipomatous tumors or well-differentiated liposarcomas with fat-suppressed T1w contrast enhanced MR images available were retrospectively enrolled. Based on a Pearson's correlation coefficient threshold at 0.8, 55 out of 87 (63.2%) radiomic features were considered reproducible. Further introduction of relevancy finally selected 35 radiomic features to be integrated in the model. To predict malignant tumors, model diagnostic performances were as follow: AUROC = 0.96; sensitivity = 100%; specificity = 90%; positive predictive value = 90.9%; negative predictive value = 100% and overall accuracy = 95.0%. CONCLUSION: This work demonstrates that radiomics allows to predict malignancy in soft tissue lipomatous tumors with routinely used MR acquisition in clinical oncology. These encouraging results need to be further confirmed in an external validation population.

[Prediction of chemotherapy response in primary osteosarcoma using the machine learning technique on radiomic data]. / Prédiction de la réponse à la chimiothérapie des ostéosarcomes à partir des données radiomiques issues des IRM diagnostiques.

INTRODUCTION: Osteosarcoma is the most common malignant bone tumor before 25 years of age. Response to neoadjuvant chemotherapy determines continuation of treatment and is also a powerful prognostic factor. There are currently no reliable ways to evaluate it early. The aim is to develop a method to predict the chemotherapy response using radiomics from pre-treatment MRI. METHODS: Clinical characteristics and MRI of patients treated for local or metastatic osteosarcoma were collected retrospectively in the Rhône-Alpes region, from 2007 to 2016. On initial MRI exams, each tumor was segmented by expert radiologist and 87 radiomic features were extracted automatically. Univariate analysis was performed to assess each feature's association with histological response following neoadjuvante chemotherapy. To distinguish good histological responder from poor, we built predictive models based on support vector machines. Their classification performance was assessed with the area under operating characteristic curve receiver (AUROC) from test data. RESULTS: The analysis focused on the MRIs of 69 patients, 55.1% (38/69) of whom were good histological responders. The model obtained by support vector machines from initial MRI radiomic data had an AUROC of 0.98, a sensitivity of 100% (IC 95% [100%-100%]) and specificity of 86% (IC 95% [59.7%-111%]). DISCUSSION: Radiomic based on MRI data would predict the chemotherapy response before treatment initiation, in patients treated for osteosarcoma.

3D Chemical Shift-Encoded MRI for Volume and Composition Quantification of Abdominal Adipose Tissue During an Overfeeding Protocol in Healthy Volunteers.

BACKGROUND: Overweight and obesity are major worldwide health concerns characterized by an abnormal accumulation of fat in adipose tissue (AT) and liver. PURPOSE: To evaluate the volume and the fatty acid (FA) composition of the subcutaneous adipose tissue (SAT) and the visceral adipose tissue (VAT) and the fat content in the liver from 3D chemical-shift-encoded (CSE)-MRI acquisition, before and after a 31-day overfeeding protocol. STUDY TYPE: Prospective and longitudinal study. SUBJECTS: Twenty-one nonobese healthy male volunteers. FIELD STRENGTH/SEQUENCE: A 3D spoiled-gradient multiple echo sequence and STEAM sequence were performed at 3T. ASSESSMENT: AT volume was automatically segmented on CSE-MRI between L2 to L4 lumbar vertebrae and compared to the dual-energy X-ray absorptiometry (DEXA) measurement. CSE-MRI and MR spectroscopy (MRS) data were analyzed to assess the proton density fat fraction (PDFF) in the liver and the FA composition in SAT and VAT. Gas chromatography-mass spectrometry (GC-MS) analyses were performed on 13 SAT samples as a FA composition countermeasure. STATISTICAL TESTS: Paired t-test, Pearson's correlation coefficient, and Bland-Altman plots were used to compare measurements. RESULTS: SAT and VAT volumes significantly increased (P < 0.001). CSE-MRI and DEXA measurements were strongly correlated (r = 0.98, P < 0.001). PDFF significantly increased in the liver (+1.35, P = 0.002 for CSE-MRI, + 1.74, P = 0.002 for MRS). FA composition of SAT and VAT appeared to be consistent between localized-MRS and CSE-MRI (on whole segmented volume) measurements. A significant difference between SAT and VAT FA composition was found (P < 0.001 for CSE-MRI, P = 0.001 for MRS). MRS and CSE-MRI measurements of the FA composition were correlated with the GC-MS results (for ndb: rMRS/GC-MS = 0.83 P < 0.001, rCSE-MRI/GC-MS = 0.84, P = 0.001; for nmidb: rMRS/GC-MS = 0.74, P = 0.006, rCSE-MRI/GC-MS = 0.66, P = 0.020) DATA CONCLUSION: The follow-up of liver PDFF, volume, and FA composition of AT during an overfeeding diet was demonstrated through different methods. The CSE-MRI sequence associated with a dedicated postprocessing was found reliable for such quantification. LEVEL OF EVIDENCE: 1 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;49:1587-1599.

In vitro distinction between proinflammatory and antiinflammatory macrophages with gadolinium-liposomes and ultrasmall superparamagnetic iron oxide particles at 3.0T.

BACKGROUND: Inflammation involves a heterogeneous macrophage population, for which there is no readily available MR assessment method. PURPOSE: To assess the feasibility of distinguishing proinflammatory M1 and antiinflammatory M2 macrophages at MRI enhanced with gadolinium liposomes or ultrasmall superparamagnetic iron oxide particles. STUDY TYPE: In vitro. SPECIMEN: We employed cultured RAW macrophages. M0 macrophages were polarized with lipopolysaccharide (LPS) or interleukin-4 (IL-4), resulting in M1 or M2 macrophages. The macrophages were incubated with gadolinium (±rhodamine) liposomes or iron oxide particles and cell pellets were prepared for MRI. FIELD STRENGTH/SEQUENCE: Transverse relaxation rates and quantitative susceptibility were obtained at 3.0T with multiecho turbo spin echo and spoiled gradient echo sequences. ASSESSMENT: MRI results were compared with confocal microscopy, flow cytometry, and expression of endocytosis, M1 and M2 genes. STATISTICAL TESTS: Mann-Whitney and Kruskal-Wallis tests were performed. RESULTS: Higher transverse relaxation rates and susceptibility were observed in M1 than in M2 and M0 macrophages (P < 0.01 both with liposomes and USPIO) and significantly different susceptibility in M2 and M0 macrophages (P < 0.01 both with liposomes and USPIO). These MRI results were confirmed at confocal microscopy and flow cytometry. LPS macrophages displayed M1 gene expression, whereas IL-4 macrophages showed M2 polarization and lower endocytosis gene expression rates. DATA CONCLUSION: These in vitro results show that it is feasible to distinguish between proinflammatory M1 and antiinflammatory M2 macrophages according to their level of contrast agent uptake at MRI. LEVEL OF EVIDENCE: 1 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2019;49:1166-1173.

3T chemical shift-encoded MRI: Detection of altered proximal femur marrow adipose tissue composition in glucocorticoid users and validation with magnetic resonance spectroscopy.

BACKGROUND: Osteoporosis (OP) results in weak bone and can ultimately lead to fracture. Drugs such as glucocorticoids can also induce OP (glucocorticoid-induced osteoporosis [GIO]). Bone marrow adipose tissue composition and quantity may play a role in OP pathophysiology, but has not been thoroughly studied in GIO compared to primary OP. PURPOSE/HYPOTHESIS: Chemical shift-encoded (CSE) MRI allows detection of subregional differences in bone marrow adipose tissue composition and quantity in the proximal femur of GIO compared to OP subjects and has high agreement with the reference standard of magnetic resonance spectroscopy (MRS). STUDY TYPE: Prospective. SUBJECTS: In all, 18 OP and 13 GIO subjects. FIELDS STRENGTH: 3T. SEQUENCE: Multiple gradient-echo, stimulated echo acquisition mode (STEAM). ASSESSMENT: Subjects underwent CSE-MRI in the proximal femurs, and for each parametric map regions of interest (ROIs) were assessed in the femoral head (fHEAD), femoral neck (fNECK), Ward's triangle (fTRIANGLE), and the greater trochanter (GTROCH). In addition, we compared CSE-MRI against the reference standard of MRS performed in the femoral neck and Ward's triangle. STATISTICAL TESTS: Differences between OP/GIO were investigated using the Mann-Whitney nonparametric test. Bland-Altman methodology was used to assess measurement agreement between CSE-MRI and MRS. RESULTS: GIO compared with OP subjects demonstrated: decreased monounsaturated fat fraction (MUFA) (-2.1%, P < 0.05) in fHEAD; decreased MUFA (-3.8%, P < 0.05), increased saturated fat fraction (SFA) (5.5%, P < 0.05), and decreased T2* (-3.8 msec, P < 0.05) in fNECK; decreased proton density fat fraction (PDFF) (-15.1%, P < 0.05), MUFA (-9.8%, P < 0.05), polyunsaturated fat fraction (PUFA) (-1.8%, P < 0.01), increased SFA (11.6%, P < 0.05), and decreased T2* (-5.4 msec, P < 0.05) in fTRIANGLE; and decreased T2* (-1.5 msec, P < 0.05) in GTROCH. There was high measurement agreement between MRI and MRS using the Bland-Altman test. DATA CONCLUSION: 3T CSE-MRI may allow reliable assessment of subregional bone marrow adipose tissue (bMAT) quantity and composition in the proximal femur in a clinically reasonable scan time. Glucocorticoids may alter the lipid profile of bMAT and potentially result in reduced bone quality. LEVEL OF EVIDENCE: 2 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;50:490-496.

Chemical shift-encoded MRI for assessment of bone marrow adipose tissue fat composition: Pilot study in premenopausal versus postmenopausal women.

OBJECT: To quantify and compare subregional proximal femur bone marrow fat composition in premenopausal and postmenopausal women using chemical shift-encoded-MRI (CSE-MRI). MATERIALS AND METHODS: A multi gradient-echo sequence at 3 T was used to scan both hips of premenopausal (n = 9) and postmenopausal (n = 18) women. Subregional fat composition (saturation, poly-unsaturation, mono-unsaturation) was quantitatively assessed in the femoral head, femoral neck, Ward's triangle, greater trochanter, and proximal shaft in bone marrow adipose tissue and separately within red and yellow marrow adipose tissue. RESULTS: Significant differences in fat composition in postmenopausal compared to premenopausal women, which varied depending on the subregion analyzed, were found. Within both whole and yellow marrow adipose tissue, postmenopausal women demonstrated higher saturation (+14.7% to +43.3%), lower mono- (-11.4% to -33%) and polyunsaturation (-52 to -83%) (p < 0.05). Within red marrow adipose tissue, postmenopausal women demonstrated lower fat quantity (-16% to -24%) and decreased polyunsaturation (-80 to -120%) in the femoral neck, greater trochanter, and Ward's triangle (p < 0.05). CONCLUSION: CSE-MRI can be used to detect subregional differences in proximal femur marrow adipose tissue composition between pre- and post-menopausal women in clinically feasible scan times.

Quantification of hepatic perfusion and hepatocyte function with dynamic gadoxetic acid-enhanced MRI in patients with chronic liver disease.

The purpose of the present study was to develop and perform initial validation of dynamic MRI enhanced with gadoxetic acid as hepatobiliary contrast agent to quantify hepatic perfusion and hepatocyte function in patients with chronic liver disease. Free-breathing, dynamic gadoxetic acid-enhanced MRI was performed at 3.0 T using a 3D time-resolved angiography sequence with stochastic trajectories during 38 min. A dual-input three-compartment model was developed to derive hepatic perfusion and hepatocyte function parameters. Method feasibility was assessed in 23 patients with biopsy-proven chronic liver disease. Parameter analysis could be performed in 21 patients (91%). The hepatocyte function parameters were more discriminant than the perfusion parameters to differentiate between patients with minimal fibrosis (METAVIR F0-F1), intermediate fibrosis (F2-F3) and cirrhosis (F4). The areas under the receiver operating characteristic curves (ROCs) to diagnose significant fibrosis (METAVIR F ≥ 2) were: 0.95 (95% CI: 0.87-1; P<0.001) for biliary efflux, 0.88 (95% CI: 0.73-1; P<0.01) for sinusoidal backflux, 0.81 (95% CI: 0.61-1; P<0.05) for hepatocyte uptake fraction and 0.75 (95% CI: 0.54-1; P<0.05) for hepatic perfusion index (HPI), respectively. These initial results in patients with chronic liver diseases show that simultaneous quantification of hepatic perfusion and hepatocyte function is feasible with free breathing dynamic gadoxetic acid-enhanced MRI. Hepatocyte function parameters may be relevant to assess liver fibrosis severity.

Renal parenchyma impairment characterization in partial unilateral ureteral obstruction in mice with intravoxel incoherent motion-MRI.

Ureteropelvic junction obstruction constitutes a major cause of progressive pediatric renal disease. The biological mechanisms underlying the renal response to obstruction can be investigated using a clinically relevant mouse model of partial unilateral ureteral obstruction (pUUO). Renal function and kidney morphology data can be evaluated using renal ultrasound, scintigraphy and uro-magnetic resonance imaging (uro-MRI), but these methods are poorly linked to histological change and not all are quantitative. Here, we propose to investigate pUUO for the first time using an intravoxel incoherent motion diffusion sequence. The aim of this study was to quantitatively characterize impairment of the kidney parenchyma in the pUUO model. This quantitative MRI method was able to assess the perfusion and microstructure of the kidney without requiring the injection of a contrast agent. The results suggest that a perfusion fraction (f) reduction is associated with a decrease in the volume of the renal parenchyma, which could be related to decreased renal vascularization. The latter may occur before impairment by fibrosis and the findings are in accordance with the literature using the UUO mice model and, more specifically, on pUUO. Further investigation is required before this technique can be made available for the diagnosis and management of children with antenatal hydronephrosis and to select the optimal timing of surgery if required.

Liver Perfusion Modifies Gd-DTPA and Gd-BOPTA Hepatocyte Concentrations Through Transfer Clearances Across Sinusoidal Membranes.

BACKGROUND AND OBJECTIVES: Gadobenate dimeglumine (Gd-BOPTA) is a commercialised hepatobiliary contrast agent used during liver magnetic resonance imaging (MRI) to detect liver diseases. It enters into human hepatocytes through organic anion transporting polypeptides (OATP1B1/B3) and crosses the canalicular transporter multiple resistance-associated protein 2 (MRP2) to be excreted into bile canaliculi. Gd-BOPTA can return to sinusoids via the sinusoidal transporters MRP3/MRP4. Hepatocyte concentrations of Gd-BOPTA depend on three clearances: the sinusoidal clearance or volume of sinusoidal blood cleared of drugs per unit of time and two hepatocyte clearances (into bile canaliculi or back to sinusoids) or volume of hepatocytes cleared of drugs per unit of time in the respective liver compartments. The present study investigates whether changing liver blood flow modifies hepatocyte concentrations when plasma concentrations do not change. METHODS: We perfused normal rat livers at various portal flow rates (24, 30, and 36 ml/min) with 200 µM Gd-BOPTA and measured sinusoidal clearances, hepatocyte clearances, and hepatocyte concentrations of Gd-BOPTA. RESULTS: We showed that varying portal flow rates changes the sinusoidal clearance of Gd-BOPTA despite its low extraction ratio. Portal flow rates do not modify Gd-BOPTA clearance from hepatocytes into bile canaliculi but can change hepatocyte clearance back to sinusoids. CONCLUSION: At a given perfused concentration, portal flow rates modify Gd-BOPTA hepatocyte concentrations, a result important to consider when interpreting liver imaging.

Strong antineoplastic effects of metformin in preclinical models of liver carcinogenesis.

Studies suggest that metformin, widely used for treating Type 2 diabetes, possesses innate antineoplastic properties. For metabolic syndrome patients with hepatocellular carcinoma (HCC), metformin may provide antitumoral effects. We evaluated the impact of metformin on tumour growth and visceral fat composition using relevant preclinical models of metabolic syndrome. Studies were performed in three hepatoma cell lines, in HepG2 xenograft mice fed with standard chow (SC) diet, 60% high-fat diet (HFD) or 30% fructose diet (FR), and an ex vivo model of human cultured HCC slices. Visceral fatty acid composition was analysed by magnetic resonance imaging (MRI). Metformin had a dose-dependent inhibitory effect on cell proliferation and apoptosis in vitro through the deregulation of mTOR/AMPK, AKT and extracellular signal regulated kinase (ERK) signalling pathways. Tumour engraftment rates were higher in HFD mice than SC mice (hepatic: 79% compared with 25%, P=0.02) and FR mice (subcutaneous: 86% compared with 50%, P=0.04). Subcutaneous tumour volume was increased in HFD mice (+64% compared with FR and SC, P=0.03). Metformin significantly decreased subcutaneous tumour growth via cell-cycle block and mammalian target of rapamycin (mTOR) pathway inhibition, and also induced hypoxia and decreased angiogenesis. In ex vivo tumour slices, metformin treatment led to increased necrosis, decreased cyclin D1 and increased carbonic anhydrase-9 (CA-9). Metformin caused qualitative changes in visceral fat composition of HFD mice, with decreased proportions of polyunsaturated fatty acids (14.6% ± 2.3% compared with 17.9% ± 3.0%, P=0.04). The potent antitumoral effects of metformin in multiple preclinical models implicating several molecular mechanisms provide a strong rationale for clinical trials including combination studies in HCC patients.

Enabling 3D-Liver Perfusion Mapping from MR-DCE Imaging Using Distributed Computing.

An MR acquisition protocol and a processing method using distributed computing on the European Grid Infrastructure (EGI) to allow 3D liver perfusion parametric mapping after Magnetic Resonance Dynamic Contrast Enhanced (MR-DCE) imaging are presented. Seven patients (one healthy control and six with chronic liver diseases) were prospectively enrolled after liver biopsy. MR-dynamic acquisition was continuously performed in free-breathing during two minutes after simultaneous intravascular contrast agent (MS-325 blood pool agent) injection. Hepatic capillary system was modeled by a 3-parameters one-compartment pharmacokinetic model. The processing step was parallelized and executed on the EGI. It was modeled and implemented as a grid workflow using the Gwendia language and the MOTEUR workflow engine. Results showed good reproducibility in repeated processing on the grid. The results obtained from the grid were well correlated with ROI-based reference method ran locally on a personal computer. The speed-up range was 71 to 242 with an average value of 126. In conclusion, distributed computing applied to perfusion mapping brings significant speed-up to quantification step to be used for further clinical studies in a research context. Accuracy would be improved with higher image SNR accessible on the latest 3T MR systems available today.