Performance degrades less under increased workload with the addition of speech control in a dynamic environment.

This research empirically evaluates the introduction of speech to existing keyboard and mouse input modalities in an application used to control aircraft in a simulated, complex and dynamic environment. Task performance and task performance degradation are assessed for three levels of workload. Previous studies have evaluated task performance using these modalities however, only a couple have evaluated task performance under varying workload. Even though speech is a common addition to modern control interfaces, the effect of varying workload on this combination of control modalities has not yet been reported. Thirty-six participants commanded simulated aircraft through generated obstacle courses to reach a Combat Air Patrol (CAP) point while also responding to a secondary task. There were nine conditions that varied the control modality (Keyboard and Mouse (KM), Voice (V), and Keyboard, Mouse and Voice (KMV)), and workload by varying the number of aircraft being controlled (low, medium and high). Results showed that KM outperformed KMV and V for the low and medium workload levels. However, task performance with KMV was found to degrade the least as workload increased. KMV and KM were found to enable significantly more correct responses to the secondary task which was delivered aurally. Participants reported a preference for the combined modalities (KMV), self-assessing that KMV most reduced their workload. This research suggests that the addition of a speech interface to existing keyboard and mouse modalities, for control of aircraft in a simulation, may help manage cognitive load and may assist in controlling more aircraft under higher workloads.

Incidental detection of prostate cancer with computed tomography scans.

Prostate cancer (PCa) is the second most frequent type of cancer found in men worldwide, with around one in nine men being diagnosed with PCa within their lifetime. PCa often shows no symptoms in its early stages and its diagnosis techniques are either invasive, resource intensive, or has low efficacy, making widespread early detection onerous. Inspired by the recent success of deep convolutional neural networks (CNN) in computer aided detection (CADe), we propose a new CNN based framework for incidental detection of clinically significant prostate cancer (csPCa) in patients who had a CT scan of the abdomen/pelvis for other reasons. While CT is generally considered insufficient to diagnose PCa due to its inferior soft tissue characterisation, our evaluations on a relatively large dataset consisting of 139 clinically significant PCa patients and 432 controls show that the proposed deep neural network pipeline can detect csPCa patients at a level that is suitable for incidental detection. The proposed pipeline achieved an area under the receiver operating characteristic curve (ROC-AUC) of 0.88 (95% Confidence Interval: 0.86-0.90) at patient level csPCa detection on CT, significantly higher than the AUCs achieved by two radiologists (0.61 and 0.70) on the same task.