On using machine learning algorithms to define clinically meaningful patient subgroups

Knevel, Rachel; Huizinga, Tom W J

doi:10.1136/annrheumdis-2019-215959

Cited by 5 publications

(3 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, financial services providers are increasingly relying on machine learning-based approaches for subgroup discovery in shaping investment and marketing strategies [1,2], designing bespoke portfolio and insurance products [3,4], managing risk [5], detecting fraud [6], and complying with anti-discrimination or fairness regulations [7]. In healthcare, attempts have been made in identifying and classifying patients' subgroups and clusters with similar prognoses and manifestations as well as responses to different treatment regimes in an attempt at personalized medical service provision [8][9][10][11]. A key objective in these tasks is the discovery of customer or patients' segments, clusters or subgroups in individual-level data, defined by demographic, psychographic, behavioral, or other variables, that are interesting or anomalous according to some criterion [12].…”

Section: Introductionmentioning

confidence: 99%

Preservation of Anomalous Subgroups On Variational Autoencoder Transformed Data

Maina¹,

Bryant²,

Ogallo³

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

In this paper, we investigate the effect of machine learning based anonymization on anomalous subgroup preservation. In particular, we train a binary classifier to discover the most anomalous subgroup in a dataset by maximizing the bias between the groups predicted odds ratio from the model and observed odds ratio from the data. We then perform anonymization using a variational autoencoder (VAE) to synthesize an entirely new dataset that would ideally be drawn from the distribution of the original data. We repeat the anomalous subgroup discovery task on the new data and compare it to what was identified pre-anonymization. We evaluated our approach using publicly available datasets from the financial industry. Our evaluation confirmed that the approach was able to produce synthetic datasets that preserved a high level of subgroup differentiation as identified initially in the original dataset. Such a distinction was maintained while having distinctly different records between the synthetic and original dataset. Finally, we packed the above end-to-end process into what we call Utility Guaranteed Deep Privacy (UGDP) system. UGDP can be easily extended to onboard alternative generative approaches such as GANs to synthesize tabular data.

show abstract

Section: Introductionmentioning

confidence: 99%

Preservation of Anomalous Subgroups On Variational Autoencoder Transformed Data

Maina¹,

Bryant²,

Ogallo³

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…We thank Knevel and Huizinga for their good comments1 on our work in which we used hierarchical clustering on principal components to define clinically meaningful subgroups of patients with anti-Ku antibodies2 and, to a lesser extent, on the work of Mariampillai et al 3 in which the authors used a similar method to reveal subgroups of myositis and propose a classification of these diseases.…”

mentioning

confidence: 99%

“…Finally, Knevel and Huizinga point out the importance of the validation of the findings. Given the extreme rarity of anti-Ku patients, we used k-fold cross-validation method as detailed by the authors 1 2. Furthermore, Mariampillai et al 3 have used the independent cohort of 50 patients to confirm 4 groups of myositis they highlighted through hierarchical clustering on principal components analysis.…”

mentioning

confidence: 99%