DS Biopharma (Ireland)

ABSTRACT Purpose Establish the minimally clinically important difference (MCID) for the Orthostatic Hypotension Questionnaire (OHQ). Background Neurogenic orthostatic hypotension (nOH) causes disabling symptoms that impair daily function and quality of life. The OHQ is a validated patient reported outcome with a symptom assessment (OHSA) and daily activity scale (OHDAS), widely used in clinical trials, despite the MCID being unestablished. Methods We analyzed data from two phase 3, randomized placebo-controlled trials (SEQUOIA and REDWOOD), evaluating ampreloxetine for symptomatic nOH in patients with Parkinson disease, multiple system atrophy, and pure autonomic failure. Using anchor-based and distribution-based methods, we calculated the MCID for the total OHQ score, OHSA and OHDAS composite subscales, and for the single dizziness/lightheadedness question (OHSA1). Results The analysis included 184 subjects from SEQUOIA and 128 from REDWOOD. The total OHQ MCID for improvement was a reduction of 0.9 to 1.2 points and for worsening was an increase of 0.7 to 1.1 points. The MCID for the OHSA composite ranged from a reduction of 0.9 to 1.3 points for improvement and an increase of 0.7 to 1.1 points for worsening. For the single item OHSA1, the MCID was a reduction of 2.0 to 3.0 points for improvement and an increase of 1.0 point for worsening. Due to poor correlation with the symptom-based anchors, a reliable MCID for the OHDAS component was not established. Conclusion These MCID thresholds for the OHQ, OHSA and OHSA item 1 alone, enhance the interpretability of scores and support their use in evaluating clinical benefit.

Abstract Conservation of biodiversity requires plant species identification skills, and automatic detection is a challenging and fascinating task for both computer/data scientists and botanists alike. This chapter describes a deep learning Convolutional Neural Network (CNN), which is trained to perform mobile imagery classification on plant species found throughout Ireland. The dataset of plant-classified RGB images underwent significant pre-processing, particularly in relation to background removal and data augmentation. Several models of deep learning CNN, with varying amounts of layers and training methods, have been evaluated on this dataset. Several deep learning models were trained and evaluated to document the speed and robustness of the flora identification. The highest performing model was then embedded in a web application, creating an online system to allow for new plant images to be uploaded and classified. This chapter highlights the main research challenges associated with this work, concludes with a mobile-based application, and discusses future research.KeywordsDeep learningMachine learningImage processingPlant speciesConvolutional neural networkArtificial intelligence

BackgroundSome antiepileptic drugs (AEDs) induce expression of hepatic enzymes. This can contribute to comorbidities via interference with metabolic pathways and concomitant drug metabolization, thereby increasing the likelihood of health care interventions. Using medical records, we compared the direct health care cost in patients initiating epilepsy therapy with enzyme-inducing AEDs (EIAEDs) vs non-enzyme-active AEDs (nEAAEDs) over up to 12 years.MethodsPatients with untreated epilepsy were indexed in the UK Clinical Practice Research Datalink and Hospital Episode Statistics database when prescribed a new EIAED or nEAAED between January 2001 and December 2010. Propensity score matching reduced confounding factors between cohorts. Patients were followed until cohort treatment failure or data cut-off. The primary outcome was the median standardized monthly direct health care cost during follow-up in 2014 £GBP, calculated using published reference costs and compared using a Mann–Whitney U test.ResultsThe unmatched EIAED cohort (n = 2752) was older (54 vs 46 years), more likely to be male, had more comorbidities, and higher health care resource use/cost during the 1-year pre-index period (median £3014 vs £2516) than the nEAAED cohort (n = 2,137). The most common index EIAED and nEAAED were carbamazepine (63.3%) and lamotrigine (58.0%), respectively. After matching, cohorts had similar features (n = 951 each). Over up to 12 years of follow-up, the median standardized monthly direct health care cost was £229 for the EIAED and £188 for the nEAAED cohorts (p = 0.0091). The median cost was higher for the EIAED cohort in every year of follow-up. In the two cohorts, 25.1% and 20.1% of total mean cost during follow-up was epilepsy-related, with approximately 4.6% and 3.0% for AED acquisition, respectively. The median time to cohort treatment failure was shorter in the matched EIAED cohort (468 vs 1194 days).ConclusionsPatients in the UK who initiated epilepsy therapy with an EIAED appeared to be at higher risk of complications associated with enzyme induction. In long-term matched cohort analyses, the median total direct health care cost associated with EIAED therapy was higher than with nEAAEDs. Changing current treatment practices could potentially improve patient outcomes and reduce costs.