Fully automated contrast selection of joint bright- and black-blood late gadolinium enhancement imaging for robust myocardial scar assessment.

PURPOSE: Joint bright- and black-blood MRI techniques provide improved scar localization and contrast. Black-blood contrast is obtained after the visual selection of an optimal inversion time (TI) which often results in uncertainties, inter- and intra-observer variability and increased workload. In this work, we propose an artificial intelligence-based algorithm to enable fully automated TI selection and simplify myocardial scar imaging. METHODS: The proposed algorithm first localizes the left ventricle using a U-Net architecture. The localized left cavity centroid is extracted and a squared region of interest ("focus box") is created around the resulting pixel. The focus box is then propagated on each image and the sum of the pixel intensity inside is computed. The smallest sum corresponds to the image with the lowest intensity signal within the blood pool and healthy myocardium, which will provide an ideal scar-to-blood contrast. The image's corresponding TI is considered optimal. The U-Net was trained to segment the epicardium in 177 patients with binary cross-entropy loss. The algorithm was validated retrospectively in 152 patients, and the agreement between the algorithm and two magnetic resonance (MR) operators' prediction of TI values was calculated using the Fleiss' kappa coefficient. Thirty focus box sizes, ranging from 2.3mm2 to 20.3cm2, were tested. Processing times were measured. RESULTS: The U-Net's Dice score was 93.0 ± 0.1%. The proposed algorithm extracted TI values in 2.7 ± 0.1 s per patient (vs. 16.0 ± 8.5 s for the operator). An agreement between the algorithm's prediction and the MR operators' prediction was found in 137/152 patients (κ= 0.89), for an optimal focus box of size 2.3cm2. CONCLUSION: The proposed fully-automated algorithm has potential of reducing uncertainties, variability, and workload inherent to manual approaches with promise for future clinical implementation for joint bright- and black-blood MRI.

Prognosis of walking function in multiple sclerosis supported by gait pattern analysis.

BACKGROUND: Walking impairment is a common and highly disabling symptom in people with MS (PwMS). Ambulatory deterioration is poorly characterized in PwMS and reliable prognosis that may guide clinical decisions is elusive. This study aimed to objectively track the progression of clinical walking performance and kinematic gait patterns in PwMS over 4 years, thereby revealing potential prognostic markers for deterioration of ambulatory function. METHODS: Twenty-two PwMS (48.8 ± 9.9 years, 14 females; expanded disability status scale [EDSS]: 4.5 ± 0.9 points) with gait impairments were recruited at the University Hospital Zurich, Switzerland. Gait function was monitored over a period of 4 years using a set of standardized clinical walking tests (timed 25-foot walk [T25FW], 6 min walk test [6MWT], 12-item MS walking scale [MSWS-12]) and comprehensive 3D kinematic gait analysis. Walking decline was assessed in the full patient cohort and in patient sub-groups that were built according to MS type (relapsing-remitting [RRMS], progressive [PMS]) and subjects' pathological gait signature (cluster groups 1-3). RESULTS: In the total cohort (n = 22), we found a significant worsening in the 6MWT (BL vs. 4y: -41.1 m; P = 0.0053), while the performance in the T25FW, MSWS-12 and the EDSS remained unchanged over 4 years. Subjects with PMS (n = 12) showed a significant worsening in the EDSS (BL vs. 4y: +0.6 points; P = 0.0053), which was not observed in participants with RRMS (n = 10). Whereas deterioration of clinical walking function was not different between subjects with RRMS and PMS, we identified differences in clinical walking deterioration between PwMS with varying gait pattern pathologies: Subjects with spastic-paretic gait impairments (cluster 1; n = 9) demonstrated a marked worsening in the T25FW (BL vs. 4y: +2 s; P = 0.0020) and 6MWT (BL vs. 4y: -92.9 m; P < 0.0001) which was not seen in PwMS with an ataxia-like (cluster 2; n = 8) or unstable walking pattern (cluster 3; n = 5). Deterioration of clinical walking performance in cluster 1 was accompanied by a specific worsening of gait deficits that were characteristic of this cluster at baseline, a phenomenon not found in the other sub-groups. Accordingly, aggravation of cluster 1-specific gait impairments over 4 years predicted deterioration of the 6MWT in the total cohort (n = 22) with an accuracy of 90.9% (sensitivity: 90.9%; specificity: 90.9%; Nagelkerkes coefficient of determination R2: 0.721), unveiling key determinants of MS-related walking decline. CONCLUSIONS: Our findings highlight the potential of quantitative, functional outcomes for objective tracking of disease progression in PwMS. Gait pattern analysis can provide valuable information on the underlying pathomechanisms of gait deterioration and may represent a complementary prognostic tool for walking function in PwMS. CLINICAL TRIAL: clinicaltrials.gov, NCT01576354.

Combining rescanning and gating for a time-efficient treatment of mobile tumors using pencil beam scanning proton therapy.

BACKGROUND AND PURPOSE: Respiratory motion during proton therapy can severely degrade dose distributions, particularly due to interplay effects when using pencil beam scanning. Combined rescanning and gating treatments for moving tumors mitigates dose degradation, but at the cost of increased treatment delivery time. The objective of this study was to identify the time efficiency of these dose degradation-motion mitigation strategies for different range of motions. MATERIALS AND METHODS: Seventeen patients with thoracic or abdominal tumors were studied. Tumor motion amplitudes ranged from 2-30 mm. Deliveries using different combinations of rescanning and gating were simulated with a dense dose spot grid (4 × 4 × 2.5 mm3) for all patients and a sparse dose spot grid (8 × 8 × 5 mm3) for six patients with larger tumor movements (>8 mm). The resulting plans were evaluated in terms of CTV coverage and time efficiency. RESULTS: Based on the studied patient cohort, it has been shown that for amplitudes up to 5 mm, no motion mitigation is required with a dense spot grid. For amplitudes between 5 and 10 mm, volumetric rescanning should be applied while maintaining a 100% duty cycle when using a dense spot grid. Although gating could be envisaged to reduce the target volume for intermediate motion, it has been shown that the dose to normal tissues would only be reduced marginally. Moreover, the treatment time would increase. Finally, for larger motion amplitudes, both volumetric rescanning and respiratory gating should be applied with both spot grids. In addition, it has been shown that a dense spot grid delivers better CTV dose coverage than a sparse dose grid. CONCLUSION: Volumetric rescanning and/or respiratory gating can be used in order to effectively and efficiently mitigate dose degradation due to tumor movement.