| 1. |
- Botvinik-Nezer, Rotem, et al.
(författare)
-
Variability in the analysis of a single neuroimaging dataset by many teams
- 2020
-
Ingår i: Nature. - : Springer Science and Business Media LLC. - 0028-0836 .- 1476-4687. ; 582, s. 84-88
-
Tidskriftsartikel (refereegranskat)abstract
- Data analysis workflows in many scientific domains have become increasingly complex and flexible. Here we assess the effect of this flexibility on the results of functional magnetic resonance imaging by asking 70 independent teams to analyse the same dataset, testing the same 9 ex-ante hypotheses(1). The flexibility of analytical approaches is exemplified by the fact that no two teams chose identical workflows to analyse the data. This flexibility resulted in sizeable variation in the results of hypothesis tests, even for teams whose statistical maps were highly correlated at intermediate stages of the analysis pipeline. Variation in reported results was related to several aspects of analysis methodology. Notably, a meta-analytical approach that aggregated information across teams yielded a significant consensus in activated regions. Furthermore, prediction markets of researchers in the field revealed an overestimation of the likelihood of significant findings, even by researchers with direct knowledge of the dataset(2-5). Our findings show that analytical flexibility can have substantial effects on scientific conclusions, and identify factors that may be related to variability in the analysis of functional magnetic resonance imaging. The results emphasize the importance of validating and sharing complex analysis workflows, and demonstrate the need for performing and reporting multiple analyses of the same data. Potential approaches that could be used to mitigate issues related to analytical variability are discussed. The results obtained by seventy different teams analysing the same functional magnetic resonance imaging dataset show substantial variation, highlighting the influence of analytical choices and the importance of sharing workflows publicly and performing multiple analyses.
|
|
| 2. |
- Belov, Vladimir, et al.
(författare)
-
Multi-site benchmark classification of major depressive disorder using machine learning on cortical and subcortical measures
- 2024
-
Ingår i: Scientific Reports. - : NATURE PORTFOLIO. - 2045-2322. ; 14:1
-
Tidskriftsartikel (refereegranskat)abstract
- Machine learning (ML) techniques have gained popularity in the neuroimaging field due to their potential for classifying neuropsychiatric disorders. However, the diagnostic predictive power of the existing algorithms has been limited by small sample sizes, lack of representativeness, data leakage, and/or overfitting. Here, we overcome these limitations with the largest multi-site sample size to date (N = 5365) to provide a generalizable ML classification benchmark of major depressive disorder (MDD) using shallow linear and non-linear models. Leveraging brain measures from standardized ENIGMA analysis pipelines in FreeSurfer, we were able to classify MDD versus healthy controls (HC) with a balanced accuracy of around 62%. But after harmonizing the data, e.g., using ComBat, the balanced accuracy dropped to approximately 52%. Accuracy results close to random chance levels were also observed in stratified groups according to age of onset, antidepressant use, number of episodes and sex. Future studies incorporating higher dimensional brain imaging/phenotype features, and/or using more advanced machine and deep learning methods may yield more encouraging prospects.
|
|
| 3. |
- Chow, Clara K., et al.
(författare)
-
TEXT messages to improve MEDication adherence and Secondary prevention (TEXTMEDS) after acute coronary syndrome: a randomised clinical trial protocol
- 2018
-
Ingår i: BMJ Open. - : BMJ PUBLISHING GROUP. - 2044-6055. ; 8:1
-
Tidskriftsartikel (refereegranskat)abstract
- Background Identifying simple, low-cost and scalable means of supporting lifestyle change and medication adherence for patients following a cardiovascular (CV) event is important. Objective The TEXTMEDS (TEXT messages to improve MEDiGation adherence and Secondary prevention) study aims to investigate whether a cardiac education and support programme sent via mobile phone text message improves medication adherence and risk factor levels in patients following an acute coronary syndrome (ACS). Study design A single-blind, multicentre, randomised clinical trial of 1400 patients after an ACS with 12 months follow-up. The intervention group will receive multiple weekly text messages that provide information, motivation, support to adhere to medications, quit smoking (if relevant) and recommendations for healthy diet and exercise. The primary endpoint is the percentage of patients who are adherent to cardioprotective medications and the key secondary outcomes are mean systolic blood pressure (BP) and low density lipoprotein cholesterol. Secondary outcomes will also include total cholesterol, mean diastolic BP, the percentage of participants who are adherent to each cardioprotective medication class, the percentage of participants who achieve target levels of CV risk factors, major vascular events, hospital readmissions and all-cause mortality. The study will be augmented by formal economic and proGess evaluations to assess acceptability, utility and Gost-effectiveness. Summary The study will provide multicentre randomised trial evidence of the effects of a text message-based programme on cardioprotective medication adherence and levels of CV risk factors. Ethics and dissemination Primary ethics approval was received from Western Sydney Local Health District Human Research EthiGs Committee (HREC2012/12/4.1 (3648) AU RED HREC/13ANMEAD/15). Results will be disseminated via peer-reviewed publications and presentations at international conferences.
|
|
