Effectiveness of Community versus Hospital Eye Service follow-up for patients with neovascular age-related macular degeneration with quiescent disease (ECHoES): a virtual non-inferiority trial

OBJECTIVES: To compare the ability of ophthalmologists versus optometrists to correctly classify retinal lesions due to neovascular age-related macular degeneration (nAMD).

DESIGN: Randomised balanced incomplete block trial. Optometrists in the community and ophthalmologists in the Hospital Eye Service classified lesions from vignettes comprising clinical information, colour fundus photographs and optical coherence tomographic images. Participants' classifications were validated against experts' classifications (reference standard).

SETTING: Internet-based application.

PARTICIPANTS: Ophthalmologists with experience in the age-related macular degeneration service; fully qualified optometrists not participating in nAMD shared care.

INTERVENTIONS: The trial emulated a conventional trial comparing optometrists' and ophthalmologists' decision-making, but vignettes, not patients, were assessed. Therefore, there were no interventions and the trial was virtual. Participants received training before assessing vignettes.

MAIN OUTCOME MEASURES: Primary outcome-correct classification of the activity status of a lesion based on a vignette, compared with a reference standard. Secondary outcomes-potentially sight-threatening errors, judgements about specific lesion components and participants' confidence in their decisions.

RESULTS: In total, 155 participants registered for the trial; 96 (48 in each group) completed all assessments and formed the analysis population. Optometrists and ophthalmologists achieved 1702/2016 (84.4%) and 1722/2016 (85.4%) correct classifications, respectively (OR 0.91, 95% CI 0.66 to 1.25; p=0.543). Optometrists' decision-making was non-inferior to ophthalmologists' with respect to the prespecified limit of 10% absolute difference (0.298 on the odds scale). Optometrists and ophthalmologists made similar numbers of sight-threatening errors (57/994 (5.7%) vs 62/994 (6.2%), OR 0.93, 95% CI 0.55 to 1.57; p=0.789). Ophthalmologists assessed lesion components as present less often than optometrists and were more confident about their classifications than optometrists.

CONCLUSIONS: Optometrists' ability to make nAMD retreatment decisions from vignettes is not inferior to ophthalmologists' ability. Shared care with optometrists monitoring quiescent nAMD lesions has the potential to reduce workload in hospitals.

TRIAL REGISTRATION NUMBER: ISRCTN07479761; pre-results registration.

N-opcode analysis for android malware classification and categorization

Malware detection is a growing problem particularly on the Android mobile platform due to its increasing popularity and accessibility to numerous third party app markets. This has also been made worse by the increasingly sophisticated detection avoidance techniques employed by emerging malware families. This calls for more effective techniques for detection and classification of Android malware. Hence, in this paper we present an n-opcode analysis based approach that utilizes machine learning to classify and categorize Android malware. This approach enables automated feature discovery that eliminates the need for applying expert or domain knowledge to define the needed features. Our experiments on 2520 samples that were performed using up to 10-gram opcode features showed that an f-measure of 98% is achievable using this approach.