Darryl Wright

Darryl E. Wright, Cole Cook, Jason Klug et al.

The de facto standard of dynamic histogram binning for radiomic feature extraction leads to an elevated sensitivity to fluctuations in annotated regions. This may impact the majority of radiomic studies published recently and contribute to issues regarding poor reproducibility of radiomic-based machine learning that has led to significant efforts for data harmonization; however, we believe the issues highlighted here are comparatively neglected, but often remedied by choosing static binning. The field of radiomics has improved through the development of community standards and open-source libraries such as PyRadiomics. But differences in image acquisition, systematic differences between observers' annotations, and preprocessing steps still pose challenges. These can change the distribution of voxels altering extracted features and can be exacerbated with dynamic binning.

Darryl E. Wright, Adriana V. Gregory, Deema Anaam et al.

Uterine leiomyosarcoma (LMS) is a rare but aggressive malignancy. On imaging, it is difficult to differentiate LMS from, for example, degenerated leiomyoma (LM), a prevalent but benign condition. We curated a data set of 115 axial T2-weighted MRI images from 110 patients (mean [range] age=45 [17-81] years) with UTs that included five different tumor types. These data were randomly split stratifying on tumor volume into training (n=85) and test sets (n=30). An independent second reader (reader 2) provided manual segmentations for all test set images. To automate segmentation, we applied nnU-Net and explored the effect of training set size on performance by randomly generating subsets with 25, 45, 65 and 85 training set images. We evaluated the ability of radiomic features to distinguish between types of UT individually and when combined through feature selection and machine learning. Using the entire training set the mean [95% CI] fibroid DSC was measured as 0.87 [0.59-1.00] and the agreement between the two readers was 0.89 [0.77-1.0] on the test set. When classifying degenerated LM from LMS we achieve a test set F1-score of 0.80. Classifying UTs based on radiomic features we identify classifiers achieving F1-scores of 0.53 [0.45, 0.61] and 0.80 [0.80, 0.80] on the test set for the benign versus malignant, and degenerated LM versus LMS tasks. We show that it is possible to develop an automated method for 3D segmentation of the uterus and UT that is close to human-level performance with fewer than 150 annotated images. For distinguishing UT types, while we train models that merit further investigation with additional data, reliable automatic differentiation of UTs remains a challenge.

Shefali Garg, Zhouyuan Huo, Khe Chai Sim et al.

Automatic speech recognition (ASR) systems have been shown to have large quality disparities between the language varieties they are intended or expected to recognize. One way to mitigate this is to train or fine-tune models with more representative datasets. But this approach can be hindered by limited in-domain data for training and evaluation. We propose a new way to improve the robustness of a US English short-form speech recognizer using a small amount of out-of-domain (long-form) African American English (AAE) data. We use CORAAL, YouTube and Mozilla Common Voice to train an audio classifier to approximately output whether an utterance is AAE or some other variety including Mainstream American English (MAE). By combining the classifier output with coarse geographic information, we can select a subset of utterances from a large corpus of untranscribed short-form queries for semi-supervised learning at scale. Fine-tuning on this data results in a 38.5% relative word error rate disparity reduction between AAE and MAE without reducing MAE quality.

Mike Walmsley, Lewis Smith, Chris Lintott et al.

We use Bayesian convolutional neural networks and a novel generative model of Galaxy Zoo volunteer responses to infer posteriors for the visual morphology of galaxies. Bayesian CNN can learn from galaxy images with uncertain labels and then, for previously unlabelled galaxies, predict the probability of each possible label. Our posteriors are well-calibrated (e.g. for predicting bars, we achieve coverage errors of 11.8% within a vote fraction deviation of 0.2) and hence are reliable for practical use. Further, using our posteriors, we apply the active learning strategy BALD to request volunteer responses for the subset of galaxies which, if labelled, would be most informative for training our network. We show that training our Bayesian CNNs using active learning requires up to 35-60% fewer labelled galaxies, depending on the morphological feature being classified. By combining human and machine intelligence, Galaxy Zoo will be able to classify surveys of any conceivable scale on a timescale of weeks, providing massive and detailed morphology catalogues to support research into galaxy evolution.

Lucy Fortson, Darryl Wright, Chris Lintott et al.

Over the past decade, Citizen Science has become a proven method of distributed data analysis, enabling research teams from diverse domains to solve problems involving large quantities of data with complexity levels which require human pattern recognition capabilities. With over 120 projects built reaching nearly 1.7 million volunteers, the Zooniverse.org platform has led the way in the application of Citizen Science as a method for closing the Big Data analysis gap. Since the launch in 2007 of the Galaxy Zoo project, the Zooniverse platform has enabled significant contributions across many disciplines; e.g., in ecology, humanities, and astronomy. Citizen science as an approach to Big Data combines the twin advantages of the ability to scale analysis to the size of modern datasets with the ability of humans to make serendipitous discoveries. To cope with the larger datasets looming on the horizon such as astronomy's Large Synoptic Survey Telescope (LSST) or the 100's of TB from ecology projects annually, Zooniverse has been researching a system design that is optimized for efficiency in task assignment and incorporating human and machine classifiers into the classification engine. By making efficient use of smart task assignment and the combination of human and machine classifiers, we can achieve greater accuracy and flexibility than has been possible to date. We note that creating the most efficient system must consider how best to engage and retain volunteers as well as make the most efficient use of their classifications. Our work thus focuses on understanding the factors that optimize efficiency of the combined human-machine system. This paper summarizes some of our research to date on integration of machine learning with Zooniverse, while also describing new infrastructure developed on the Zooniverse platform to carry out this research.