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.

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.

Synthesis of Lipophilic Core-Shell Fe3O4@SiO2@Au Nanoparticles and Polymeric Entrapment into Nanomicelles: A Novel Nanosystem for in Vivo Active Targeting and Magnetic Resonance-Photoacoustic Dual Imaging.

In this work, iron/silica/gold core-shell nanoparticles (Fe3O4@SiO2@Au NPs) characterized by magnetic and optical properties have been synthesized to obtain a promising theranostic platform. To improve their biocompatibility, the obtained multilayer nanoparticles have been entrapped in polymeric micelles, decorated with folic acid moieties, and tested in vivo for photoacoustic and magnetic resonance imaging detection of ovarian cancer.