An Adaptive Low-Rank Modeling-Based Active Learning Method for Medical Image Annotation

Active learning is an effective solution to interactively select a limited number of informative examples and use them to train a learning algorithm that can achieve its optimal performance for specific tasks. It is suitable for medical image applications in which unlabeled data are abundant but manual annotation could be very time-consuming and expensive. However, designing an effective active learning strategy for informative example selection is a challenging task, due to the intrinsic presence of noise in medical images, the large number of images, and the variety of imaging modalities. In this study, a novel low-rank modeling-based multi-label active learning (LRMMAL) method is developed to address these challenges and select informative examples for training a classifier to achieve the optimal performance. The proposed method independently quantifies image noise and integrates it with other measures to guide a pool-based sampling process to determine the most informative examples for training a classifier. In addition, an automatic adaptive cross entropy-based parameter determination scheme is proposed for further optimizing the example sampling strategy. Experimental results on varied medical image datasets and comparisons with other state-of-the-art multi-label active learning methods illustrate the superior performance of the proposed method.

Magnetic Resonance Image-Guided Radiotherapy (MRIgRT): A 4.5-Year Clinical Experience

AIMS

Magnetic resonance image-guided radiotherapy (MRIgRT) has been clinically implemented since 2014. This technology offers improved soft-tissue visualisation, daily imaging, and intra-fraction real-time imaging without added radiation exposure, and the opportunity for adaptive radiotherapy (ART) to adjust for anatomical changes. Here we share the longest single-institution experience with MRIgRT, focusing on trends and changes in use over the past 4.5 years.

MATERIALS AND METHODS

We analysed clinical information, including patient demographics, treatment dates, disease sites, dose/fractionation, and clinical trial enrolment for all patients treated at our institution using MRIgRT on a commercially available, integrated 0.35 T MRI, tri-cobalt-60 device from 2014 to 2018. For each patient, factors including disease site, clinical rationale for MRIgRT use, use of ART, and proportion of fractions adapted were summated and compared between individual years of use (2014-2018) to identify shifts in institutional practice patterns.

RESULTS

Six hundred and forty-two patients were treated with 666 unique treatment courses using MRIgRT at our institution between 2014 and 2018. Breast cancer was the most common disease, with use of cine MRI gating being a particularly important indication, followed by abdominal sites, where the need for cine gating and use of ART drove MRIgRT use. One hundred and ninety patients were treated using ART in 1550 fractions, 67.6% (1050) of which were adapted. ART was primarily used in cancers of the abdomen. Over time, breast and gastrointestinal cancers became increasingly dominant for MRIgRT use, hypofractionated treatment courses became more popular, and gastrointestinal cancers became the principal focus of ART.

DISCUSSION

MRIgRT is widely applicable within the field of radiation oncology and new clinical uses continue to emerge. At our institution to date, applications such as ART for gastrointestinal cancers and accelerated partial breast irradiation (APBI) for breast cancer have become dominant indications, although this is likely to continue to evolve.