Med Phys
-
Comparative Study
Machine learning algorithms for outcome prediction in (chemo)radiotherapy: An empirical comparison of classifiers.
Machine learning classification algorithms (classifiers) for prediction of treatment response are becoming more popular in radiotherapy literature. General Machine learning literature provides evidence in favor of some classifier families (random forest, support vector machine, gradient boosting) in terms of classification performance. The purpose of this study is to compare such classifiers specifically for (chemo)radiotherapy datasets and to estimate their average discriminative performance for radiation treatment outcome prediction. ⋯ Random forest and elastic net logistic regression yield higher discriminative performance in (chemo)radiotherapy outcome and toxicity prediction than other studied classifiers. Thus, one of these two classifiers should be the first choice for investigators when building classification models or to benchmark one's own modeling results against. Our results also show that an informed preselection of classifiers based on existing datasets can improve discrimination over random selection.
-
To evaluate the effectiveness of CT image-guided proton radiotherapy for prostate cancer by analyzing the positioning uncertainty and assessing daily dose change due to anatomical variations. ⋯ The correction of repositioning along the AP and SI direction from conventional bone matching in CT image-guided proton therapy was found to be effective to maintain the dose constraint of the rectum and the dose coverage of the prostate. This work indicated that prostate cancer treatment by prostate matching using CT image guidance may be effective to reduce the rectal complications and achieve better tumor control of the prostate. However, an adaptive approach is desirable to maintain better dose coverage of the SVs.
-
Myocardial infarct (MI) may consist of an infarct core (IC) and a heterogeneous, semi-viable border zone (BZ). Patients with chronic MI in the left ventricular (LV) myocardium are at increased risk of developing ventricular arrhythmias, and may therefore qualify for implantable cardioverter defibrillator (ICD) therapy. Indices based on MI mass, as determined by cardiac magnetic resonance (CMR) imaging, are shown to be sensitive in predicting adverse ventricular arrhythmic events. However, several factors, such as imaging technique and spatial resolution affect the accuracy of MI mass quantification. The aim of this study was to compare the MI masses determined by T1-mapping CMR techniques to those of conventional late Gadolinium-enhanced cardiac magnetic resonance (LGE-CMR) using inversion recovery fast gradient echo (IR-FGRE). We additionally aimed to investigate the effect of diminishing image resolution on quantification of the MI mass and its ability to predict appropriate ICD therapy. ⋯ While estimated IC masses were consistent among all three techniques, the estimated BZ masses were not consistent, especially when spatial resolution of images differed between the techniques. In particular, our study showed that diminished image resolution caused an increase in estimation of the BZ mass, likely due to partial volume effects, which led to a reduced sensitivity in the prediction of appropriate ICD therapy.
-
Cardiac perfusion PET data can be reconstructed as a dynamic sequence and kinetic modeling performed to quantify myocardial blood flow, or reconstructed as static gated images to quantify function. Parametric images from dynamic PET are conventionally not gated, to allow use of all events with lower noise. An alternative method for dynamic PET is to incorporate the kinetic model into the reconstruction algorithm itself, bypassing the generation of a time series of emission images and directly producing parametric images. So-called "direct reconstruction" can produce parametric images with lower noise than the conventional method because the noise distribution is more easily modeled in projection space than in image space. In this work, we develop direct reconstruction of cardiac-gated parametric images for 82 Rb PET with an extension of the Parametric Motion compensation OSEM List mode Algorithm for Resolution-recovery reconstruction for the one tissue model (PMOLAR-1T). ⋯ Direct reconstruction can produce parametric images with less noise than the indirect method, opening the potential utility of gated parametric imaging for perfusion PET.
-
Osteoporosis associated with reduced bone mineral density (BMD) and microarchitectural changes puts patients at an elevated risk of fracture. Modern multidetector row CT (MDCT) technology, producing high spatial resolution at increasingly lower dose radiation, is emerging as a viable modality for trabecular bone (Tb) imaging. Wide variation in CT scanners raises concerns of data uniformity in multisite and longitudinal studies. A comprehensive cadaveric study was performed to evaluate MDCT-derived Tb microarchitectural measures. A human pilot study was performed comparing continuity of Tb measures estimated from two MDCT scanners with significantly different image resolution features. ⋯ Most MDCT-derived Tb microarchitectural measures are reproducible and their values derived from two scanners strongly correlate with each other as well as with bone strength. This study has highlighted those MDCT-derived measures which show the greatest promise for characterization of bone network area density, plate-rod and transverse trabecular distributions with a good correlation (r ≥ 0.85) compared with their micro-CT-derived values. At the same time, other measures representing trabecular thickness and separation, erosion index, and structure model index produced weak correlations (r < 0.8) with their micro-CT-derived values, failing to accurately portray the projected trabecular microarchitectural features. Strong correlations of Tb measures estimated from two scanners suggest that image data from different scanners can be used successfully in multisite and longitudinal studies with linear calibration required for some measures. In summary, modern MDCT scanners are suitable for effective quantitative imaging of peripheral Tb microarchitecture if care is taken to focus on appropriate quantitative metrics.