Technical Challenges in the Clinical Application of Radiomics

Radiomics is a quantitative approach to medical image analysis targeted at deciphering the morphologic and functional features of a lesion. Radiomic methods can be applied across various malignant conditions to identify tumor phenotype characteristics in the images that correlate with their likelihood of survival, as well as their association with the underlying biology. Identifying this set of characteristic features, called tumor signature, holds tremendous value in predicting the behavior and progression of cancer, which in turn has the potential to predict its response to various therapeutic options. We discuss the technical challenges encountered in the application of radiomics, in terms of methodology, workflow integration, and user experience, that need to be addressed to harness its true potential.

Molecular Imaging in Genomic Medicine

Molecular imaging is the result of advances in the fields of molecular biology and imaging technology and has become an increasingly important tool in the discovery and understanding of a wide range of pathophysiologic processes, ranging from genetic disorders to malignant conditions. The advancement in molecular pathology techniques has enabled us to study the complex genotype of disease entities and how it impacts their behaviour and natural history. Image‐guided genomic medicine utilises methodologies to integrate genomic and radiologic data to develop insights into the genotype–phenotype relationship, which in turn can guide medical decision‐making and treatment planning

Differential diagnosis of mild cognitive impairment and Alzheimer’s disease using structural MRI cortical thickness, hippocampal shape, hippocampal texture, and volumetry

This paper presents a brain T1-weighted structural magnetic resonance imaging (MRI) biomarker that combines several individual MRI biomarkers (cortical thickness measurements, volumetric measurements, hippocampal shape, and hippocampal texture). The method was developed, trained, and evaluated using two publicly available reference datasets: a standardized dataset from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) and the imaging arm of the Australian Imaging Biomarkers and Lifestyle flagship study of ageing (AIBL). In addition, the method was evaluated by participation in the Computer-Aided Diagnosis of Dementia (CADDementia) challenge. Cross-validation using ADNI and AIBL data resulted in a multi-class classification accuracy of 62.7% for the discrimination of healthy normal controls (NC), subjects with mild cognitive impairment (MCI), and patients with Alzheimer’s disease (AD). This performance generalized to the CADDementia challenge where the method, trained using the ADNI and AIBL data, achieved a classification accuracy 63.0%. The obtained classification accuracy resulted in a first place in the challenge, and the method was significantly better (McNemar’s test) than the bottom 24 methods out of the total of 29 methods contributed by 15 different teams in the challenge. The method was further investigated with learning curve and feature selection experiments using ADNI and AIBL data. The learning curve experiments suggested that neither more training data nor a more complex classifier would have improved the obtained results. The feature selection experiment showed that both common and uncommon individual MRI biomarkers contributed to the performance; hippocampal volume, ventricular volume, hippocampal texture, and parietal lobe thickness were the most important. This study highlights the need for both subtle, localized measurements and global measurements in order to discriminate NC, MCI, and AD simultaneously based on a single structural MRI scan. It is likely that additional non-structural MRI features are needed to further improve the obtained performance, especially to improve the discrimination between NC and MCI.

Brain region’s relative proximity as marker for Alzheimer’s disease based on structural MRI

Alzheimer’s disease (AD) is a progressive, incurable neurodegenerative disease and the most common type of dementia. It cannot be prevented, cured or drastically slowed, even though AD research has increased in the past 5-10 years. Instead of focusing on the brain volume or on the single brain structures like hippocampus, this paper investigates the relationship and proximity between regions in the brain and uses this information as a novel way of classifying normal control (NC), mild cognitive impaired (MCI), and AD subjects.

Magnetic Resonance Imaging in Rheumatology


Magnetic resonance (MR) imaging has traditionally only played a small role in the clinical care of most patients with arthritis. However, with modern therapeutic strategies, early diagnosis is now more important than ever before. Consequently, advanced MRI techniques and applications now play a crucial role in managing an increasing proportion of rheumatology patients. This article reviews MR imaging techniques that are in widespread use and in development for detection and quantification of inflammation and structural damage in arthritis. It focuses on the role of MR imaging for diagnosis, management, and research in inflammatory arthropathies. Osteoarthritis and gout are briefly reviewed.

Added value of combining methotrexate with a biological agent compared to biological monotherapy in rheumatoid arthritis patients: A systematic review and meta-analysis of randomised trials


Background Methotrexate (MTX) is considered the anchor drug in rheumatoid arthritis (RA) treatment, based on efficacy, safety, and its ability to increase the efficacy of biologic agents when used in combination. Both the “American College of Rheumatology” (ACR) and the “European League Against Rheumatism” (EULAR) recommend the use of biologic agents with concomitant MTX in RA. However, analyses from health care claims suggest that the MTX prescribed in conjunction with a biologic disease-modifying antirheumatic drug (bDMARD), is not collected at the pharmacy by more than half of the patients. Despite claims from pivotal trial data, the empirical evidence supporting combination MTX plus bDMARD in terms of the actual benefit-harm has not been evaluated extensively.

Objectives To review the evidence for benefit and harm associated with combining MTX with a biologic agent in RA patients.

Methods A systematic review and meta-analysis of randomised controlled trials (RCTs) was performed to identify all trials relating treatment of RA with MTX in combination with bDMARD compared to a bDMARD in monotherapy. Random-effects models were applied for meta-analyses with risk ratio and 95% confidence intervals (RR, 95%CI) and inconsistency (I2, %) estimated using Review Manager. According to the protocol the major outcomes were ACR50 for benefit and withdrawals due to adverse events (AEs) for harm (PROSPERO: CRD42014014633).

Results From the 4,405 identified references, 14 trials were eligible for inclusion in the meta-analysis. The overall likelihood of responding to therapy (ie, RR according to ACR50) was 1.36 [1.24 to 1.49] – with a low degree of inconsistency (28%) – in favour of concomitant use of MTX when treating with a biologic (P<0.001). There was no reason to suspect that added value of MTX varied with the choice of bDMARD (Test for subgroup differences P=0.38): abatacept 1.27 [0.97 to 1.67], adalimumab 1.42 [1.24 to 1.61], etanercept 1.44 [1.22 to 1.69], golimumab 1.60 [1.13 to 2.27], infliximab 3.54 [1.38 to 9.08], rituximab 1.31 [0.74 to 2.32], and tocilizumab 1.24 [1.03 to 1.51]. The overall estimate of discontinuing therapy due to AEs from concomitant use of MTX was 1.17 [0.94 to 1.44] (P=0.16) compared to bDMARD in monotherapy – corresponding to a possible 17% increased risk.

Conclusions This systematic review and meta-analysis of RCTs provides empirical evidence about the clinical value of combining the prescribed bDMARD with the recommended concomitant use of MTX. Combination therapy increases the probability of achieving treatment success at the level of ACR50 by approximately 40% compared to patients using biologics in monotherapy. Although the precision around the estimate of concomitant MTX use does not rule out an increased risk of clinically important harms, these findings justify the recommendation that all patients prescribed biologics should be encouraged to continue MTX therapy.

Thigh Muscle MRI Reveals Extensive Muscle Edema and Early Fatty Replacement in Patients with Immune-Mediated Necrotizing Myopathy

Background/Purpose: The aim of this study was to define the pattern of muscle involvement in patients with immune-mediated necrotizing myopathy (IMNM).

Methods: All Johns Hopkins Myositis Longitudinal Cohort subjects with an available thigh magnetic resonance imaging (tMRI) who fulfilled criteria for IMNM, dermatomyositis (DM), polymyositis (PM), clinically amyopathic DM (CADM) or inclusion body myositis (IBM) were included in the study. Fifteen muscles were assessed by tMRI for the presence or absence of edema, atrophy, fatty replacement, and fascial edema. Disease subgroups were compared using univariate and multivariate analysis. Within IMNM, patients with anti-SRP and anti-HMGCR were also compared.

Results: The study included 666 subjects (101 IMNM, 176 PM, 219 DM, 17 CADM and 153 IBM). Compared to patients with DM or PM, IMNM was characterized by a higher proportion of thigh muscles with edema, atrophy, and fatty replacement (p<0.01), independent of other confounding variables. Compared with anti-SRP, anti-HMGCR patients showed a lesser proportion of muscles with atrophy (-9%, p=0.04). In IMNM, muscle abnormalities were especially common in the lateral rotator and gluteal groups. Fascial involvement was more widespread in DM. Interestingly, fatty replacement of muscle tissue began early during the course of IMNM as well as the other muscle diseases. Although it had a negative predictive value of 93%, an optimal combination of tMRI features had only a 55% positive predictive value for diagnosing IMNM.

Conclusion: Compared to patients with DM or PM, IMNM is characterized by more widespread muscle involvement. Compared with anti-HMGCR, anti-SRP patients seem to have more severe muscle involvement.

Imaging Biomarkers in Crohn’s Associated Axial Spondyloarthritis

Background/Purpose: Axial SpA has been reported in 4-16% of patients with Crohn’s disease (CD). However, some plain radiograph or magnetic resonance imaging (MRI) studies in Crohn’s population suggest presence of sacroiliitis in up to 40% patients. Whether axial SpA is underdiagnosed and hence undertreated is unclear, especially since it does not correlate with CD activity. This study utilizes standardized MRI SI joint (SIJ) imaging to determine prevalence of axial SpA in a cohort of CD patients and investigate its relationship with other clinical and serologic measures. In addition, we compare MRIs from our cohort of CD patients with separate cohorts of patients with known axial SpA with and without inflammatory bowel disease (IBD).

Methods: Adult consecutive patients meeting pathological, radiological and/or endoscopic criteria of CD and not currently on any biologic except vedolizumab, were prospectively enrolled from an IBD Clinic. Data collected included length of CD, history of joint/back pain, HLA-B27 status, BASMI, BASDAI, Harvey-Bradshaw Index (HBI) scores- a measure of CD activity, Ankylosing Spondylitis Disease Activity Score-C reactive protein (ASDAS-CRP). All patients underwent T1 and short tau inversion recovery (STIR) sequence MRI of SIJ. 3 expert readers, blinded to clinical history reviewed MRIs for presence of active and/or structural lesions globally indicative of SpA, for ASAS MRI positivity, detailed lesion scores using the SpA Research Consortium of Canada (SPARCC) SIJ inflammatory and structural scoring methods after calibration using validated modules. These images were compared to MRIs from age and sex-matched AS patients with and without IBD.

Results: 32 CD patients were enrolled: 76% females, 80% white, median age 36.4 years (IQR 27.2 – 49) with moderate CD activity (mean HBI 8.8 ± SD 4.5). 56% had peripheral SpA and only 1 was HLA-B27 positive. 20 reported back pain and 14 (70%) met ASAS inflammatory back pain (IBP) criteria. Mean BASDAI, ASDAS-CRP and BASMI were higher than patients without back pain (p <0.05). Total of 9 SIJ MRIs (35% of all patients with back pain and 16% of those without) showed global evidence of SpA (bone marrow edema BME ≥ 2 and/or structural lesions) and 6 out of those were ASAS positive (presence of active inflammation/BME), including 2 asymptomatic patients. Older age (OR 1.4, 95% CI 1.04 – 1.90) and higher BASMI (OR 3.2 95% CI 1.1-9.6) were associated with MRI SpA findings. But presence of IBP, peripheral SpA, CD duration, BASDAI, ASDAS-CRP or HBI did not show association. Compared to historic cohort of AS patients with (n=23) and without (n=24) IBD, CD patients with IBP had similar BASDAI and ASDAS-CRP scores (p <0.05) but they had lower BME and structural lesions (p 0.02 and <0.001) compared to IBD with and without SpA.

Conclusion: MRI evidence of axial SpA was 35% and 16% in our cohort of CD patients with and without back pain respectively. MRI is a valuable tool to diagnose axial SpA in CD patients with back pain, especially those who are older and has evidence of restricted spinal mobility on exam. Presence of IBP did not show association, as previously suggested, although these patients experience similar morbidity as our comparator cohort with known AS with or without IBD.