Amplifying the Effects of Contrast Agents on Magnetic Resonance Images Using a Deep Learning Method Trained on Synthetic Data.

OBJECTIVES: Artificial intelligence (AI) methods can be applied to enhance contrast in diagnostic images beyond that attainable with the standard doses of contrast agents (CAs) normally used in the clinic, thus potentially increasing diagnostic power and sensitivity. Deep learning-based AI relies on training data sets, which should be sufficiently large and diverse to effectively adjust network parameters, avoid biases, and enable generalization of the outcome. However, large sets of diagnostic images acquired at doses of CA outside the standard-of-care are not commonly available. Here, we propose a method to generate synthetic data sets to train an "AI agent" designed to amplify the effects of CAs in magnetic resonance (MR) images. The method was fine-tuned and validated in a preclinical study in a murine model of brain glioma, and extended to a large, retrospective clinical human data set. MATERIALS AND METHODS: A physical model was applied to simulate different levels of MR contrast from a gadolinium-based CA. The simulated data were used to train a neural network that predicts image contrast at higher doses. A preclinical MR study at multiple CA doses in a rat model of glioma was performed to tune model parameters and to assess fidelity of the virtual contrast images against ground-truth MR and histological data. Two different scanners (3 T and 7 T, respectively) were used to assess the effects of field strength. The approach was then applied to a retrospective clinical study comprising 1990 examinations in patients affected by a variety of brain diseases, including glioma, multiple sclerosis, and metastatic cancer. Images were evaluated in terms of contrast-to-noise ratio and lesion-to-brain ratio, and qualitative scores. RESULTS: In the preclinical study, virtual double-dose images showed high degrees of similarity to experimental double-dose images for both peak signal-to-noise ratio and structural similarity index (29.49 dB and 0.914 dB at 7 T, respectively, and 31.32 dB and 0.942 dB at 3 T) and significant improvement over standard contrast dose (ie, 0.1 mmol Gd/kg) images at both field strengths. In the clinical study, contrast-to-noise ratio and lesion-to-brain ratio increased by an average 155% and 34% in virtual contrast images compared with standard-dose images. Blind scoring of AI-enhanced images by 2 neuroradiologists showed significantly better sensitivity to small brain lesions compared with standard-dose images (4.46/5 vs 3.51/5). CONCLUSIONS: Synthetic data generated by a physical model of contrast enhancement provided effective training for a deep learning model for contrast amplification. Contrast above that attainable at standard doses of gadolinium-based CA can be generated through this approach, with significant advantages in the detection of small low-enhancing brain lesions.

An ultrasound-guided injection method for a syngeneic orthotopic murine model of breast cancer.

Breast cancer is the most common cancer among women worldwide. For high-risk women, contrast enhanced (CE)-magnetic resonance imaging (MRI) is recommended as supplemental screening together with mammography. The development of new MRI contrast agents is an active field of research, which requires efficacy tests on appropriate preclinical pathological models. In this work, a refined method to orthotopically induce breast cancer in BALB/c mice was developed using ultrasound (US) as a guide for the precise localisation of the tumour induction site and to improve animal welfare. The method was coupled with CE-MRI to characterise the evolution of the tumoural lesion.

Macrocyclic MR contrast agents: evaluation of multiple-organ gadolinium retention in healthy rats.

OBJECTIVES: The purpose of this study was to compare Gd levels in rat tissues after cumulative exposure to four commercially available macrocyclic gadolinium-based contrast agents (GBCAs). METHODS: Sixty-five male Sprague-Dawley rats were randomized to four exposure groups (n = 15 per group) and one control group (n = 5). Animals in each exposure group received 20 GBCA administrations (four per week of ProHance®, Dotarem®, Clariscan™, or Gadovist® for 5 consecutive weeks) at a dose of 0.6 mmol/kg bodyweight. After 28-days' recovery, animals were sacrificed and tissues harvested for Gd determination by inductively coupled plasma-mass spectroscopy (ICP-MS). Histologic assessment of the kidney tissue was performed for all animals. RESULTS: Significantly (p ≤ 0.005; all evaluations) lower Gd levels were noted with ProHance® than with Dotarem®, Clariscan™, or Gadovist® in all soft tissue organs: 0.144 ± 0.015 nmol/g vs. 0.342 ± 0.045, 0.377 ± 0.042, and 0.292 ± 0.047 nmol/g, respectively, for cerebrum; 0.151 ± 0.039 nmol/g vs. 0.315 ± 0.04, 0.345 ± 0.053, and 0.316 ± 0.040 nmol/g, respectively, for cerebellum; 0.361 ± 0.106 nmol/g vs. 0.685 ± 0.330, 0.823 ± 0.495, and 1.224 ± 0.664 nmol/g, respectively, for liver; 38.6 ± 25.0 nmol/g vs. 172 ± 134, 212 ± 121, and 294 ± 127 nmol/g, respectively, for kidney; and 0.400 ± 0.112 nmol/g vs. 0.660 ± 0.202, 0.688 ± 0.215, and 0.999 ± 0.442 nmol/g, respectively, for skin. No GBCA-induced macroscopic or microscopic findings were noted in the kidneys. CONCLUSIONS: Less Gd is retained in the brain and body tissues of rats 28 days after the last exposure to ProHance® compared to other macrocyclic GBCAs, likely due to unique physico-chemical features that facilitate more rapid and efficient clearance.

Orthotopic induction of CH157MN convexity and skull base meningiomas into nude mice using stereotactic surgery and MRI characterization.

Meningioma in vivo research is hampered by the difficulty of establishing an easy and reproducible orthotopic model able to mimic the characteristics of a human meningioma. Moreover, leptomeningeal dissemination and high mortality are often associated with such orthotopical models, making them useless for clinical translation studies. An optimized method for inducing meningiomas in nude mice at two different sites is described in this paper and the high reproducibility and low mortality of the models are demonstrated. Skull base meningiomas were induced in the auditory meatus and convexity meningiomas were induced on the brain surface of 23 and 24 nude mice, respectively. Both models led to the development of a mass easily observable by imaging methods. Dynamic contrast enhanced MRI was used as a tool to monitor and characterize the pathology onset and progression. At the end of the study, histology was performed to confirm the neoplastic origin of the diseased mass.

Exploiting the Proton Exchange as an Additional Route to Enhance the Relaxivity of Paramagnetic MRI Contrast Agents.

The relaxivity of Gd(HP-DO3A) was studied as a function of pH and buffer composition in order to identify the main factors of the observed relaxation enhancement due to the exchange of the coordinated hydroxyl proton. It was established that the paramagnetic relaxation time, T1M, of the coordinated hydroxyl proton is about 50% shorter than that of the protons in the coordinated water molecule. The control of the p K of the coordinated alcoholic -OH moiety in the ligand is fundamental to utilize the proton exchange enhanced relaxivity under physio/pathologic conditions. A new derivative of Gd(HP-DO3A) was synthesized by replacing the -CH3 group with a -CF3 moiety. In this complex, the -OH group becomes more acidic. Consequently, the maximum contribution of the proton exchange to the relaxivity is shifted to a lower pH region with the fluorinated ligand.

Synthesis, Characterization, and Biodistribution of a Dinuclear Gadolinium Complex with Improved Properties as a Blood Pool MRI Agent.

A dinuclear gadolinium(III) chelate containing two moieties of diethylenetriaminepentaacetic acid (DTPA), covalently conjugated to an analogue of deoxycholic acid, was synthesized and thoroughly characterized. A full relaxometric analysis was carried out, consisting of 1) the acquisition of nuclear magnetic resonance dispersion (NMRD) profiles in various media; 2) the study of binding affinity to serum albumin; 3) the measurement of 17 O transverse relaxation rate versus temperature, and 4) a transmetallation assay. In vivo biodistribution MRI studies at 1 T and blood pharmacokinetics assays were carried out in comparison with Gd-DTPA (Magnevist) and gadocoletic acid trisodium salt (B22956/1), two well-known Gd complexes that share the same chelating cage and the same deoxycholic acid residue of the Gd complex investigated herein ((GdDTPA)2 -Chol). High affinity for plasma protein and, in particular, the availability of more than one binding site, allows the complex to reach a fairly high relaxivity value in plasma (∼20 mm-1 s-1 , 20 MHz, 310 K) as well as to show unexpectedly enhanced properties of blood pooling, with an elimination half-life in rats approximately seven times longer than that of B22956/1.

Macrocyclic paramagnetic agents for MRI: Determinants of relaxivity and strategies for their improvement.

PURPOSE: To dissect the contributions to the longitudinal relaxivity (r1 ) of two commercial contrast agents (CAs), Gd-DOTA and Gd-HP-DO3A, and to synthesize/characterize a novel macrocyclic agent (Gd-Phen-DO3A) having superior r1 . METHODS: Longitudinal relaxation rates R1 of the CAs in saline with/without human serum albumin (HSA), ionized simulated body fluid (i-SBF), viscous simulated body fluid (v-SBF), and human plasma were measured. Results have been interpreted to evince the main determinants to the observed r1 values. RESULTS: In v-SBF or in the presence of HSA, r1 is enhanced for all complexes, reflecting the viscosity increase and a weak interaction with proteins. The CAs further differentiate in plasma, with a relaxivity increase (versus saline) of approximately 1, 1.5, and 2.5 mM-1 s-1 for Gd-DOTA, Gd-HPDO3A, and Gd-Phen-DO3A, respectively. R1 versus pH curves in i-SBF indicates that prototropic exchange sizably contributes to the relaxivity of Gd-HP-DO3A and Gd-Phen-DO3A. CONCLUSION: The major contributions to r1 in the physiological environment have been highlighted, namely, increased viscosity, complex-protein interaction, and prototropic exchange. The control of these terms allows the design of novel macrocyclic structures with enhanced r1 as a result of an improved interaction with plasma's macromolecules and the shift of the prototropic exchange to physiological pH. Magn Reson Med 78:1523-1532, 2017. © 2016 International Society for Magnetic Resonance in Medicine.