Clustering Clinical Data in R

Pina, Ana; Macedo, M. Paula; Henriques, Roberto

doi:10.1007/978-1-4939-9744-2_14

Cited by 9 publications

(15 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To perform a cluster analysis, impactful decisions must be made: inclusion and exclusion criteria, choice of variables and the algorithm to perform the analysis, among others. Additionally, indexes that define the best number of clusters and distance metrics have to be selected 26 . Cluster analysis used to date to tackle T2D and dysmetabolism have a dissimilar methodology that must be considered when interpreting and integrating the results (Table 2).…”

Section: Cluster Analysis Algorithm Impact On Founded Clustersmentioning

confidence: 99%

“…Furthermore, data can be analysed at different cut‐off values, allowing us to understand how observations aggregate. However, it can only find clusters with specific shapes, it gives distinct solutions depending on the chosen aggregation methodology to join the observations and has a high computation cost 26 . k ‐means is a simple and efficient algorithm.…”

Section: Cluster Analysis Algorithm Impact On Founded Clustersmentioning

confidence: 99%

“…k ‐means is a simple and efficient algorithm. Besides not dealing well with categorical variables, the final solution is highly impacted by its random initialization, requires an a priori specification of the number of clusters, and importantly, it is prone to find spherical clusters, even if this is not their natural shape 26 . The latter can limit its use.…”

Section: Cluster Analysis Algorithm Impact On Founded Clustersmentioning

confidence: 99%

“…The latter can limit its use. Partition around medoids (PAM) is a k ‐medoids algorithm, that is less sensitive to noise than k ‐means, but with a higher computational cost 26 …”

Section: Cluster Analysis Algorithm Impact On Founded Clustersmentioning

confidence: 99%

See 3 more Smart Citations

Big data and machine learning to tackle diabetes management

Pina

Meneses

Sousa‐Lima

et al. 2022

Eur J Clin Investigation

Self Cite

View full text Add to dashboard Cite

Background Type 2 Diabetes (T2D) diagnosis is based solely on glycaemia, even though it is an endpoint of numerous dysmetabolic pathways. Type 2 Diabetes complexity is challenging in a real‐world scenario; thus, dissecting T2D heterogeneity is a priority. Cluster analysis, which identifies natural clusters within multidimensional data based on similarity measures, poses a promising tool to unravel Diabetes complexity. Methods In this review, we scrutinize and integrate the results obtained in most of the works up to date on cluster analysis and T2D. Results To correctly stratify subjects and to differentiate and individualize a preventive or therapeutic approach to Diabetes management, cluster analysis should be informed with more parameters than the traditional ones, such as etiological factors, pathophysiological mechanisms, other dysmetabolic co‐morbidities, and biochemical factors, that is the millieu. Ultimately, the above‐mentioned factors may impact on Diabetes and its complications. Lastly, we propose another theoretical model, which we named the Integrative Model. We differentiate three types of components: etiological factors, mechanisms and millieu. Each component encompasses several factors to be projected in separate 2D planes allowing an holistic interpretation of the individual pathology. Conclusion Fully profiling the individuals, considering genomic and environmental factors, and exposure time, will allow the drive to precision medicine and prevention of complications.

show abstract

Section: Cluster Analysis Algorithm Impact On Founded Clustersmentioning

confidence: 99%

Section: Cluster Analysis Algorithm Impact On Founded Clustersmentioning

confidence: 99%

Section: Cluster Analysis Algorithm Impact On Founded Clustersmentioning

confidence: 99%

“…The latter can limit its use. Partition around medoids (PAM) is a k ‐medoids algorithm, that is less sensitive to noise than k ‐means, but with a higher computational cost 26 …”

Section: Cluster Analysis Algorithm Impact On Founded Clustersmentioning

confidence: 99%

See 2 more Smart Citations

Big data and machine learning to tackle diabetes management

Pina

Meneses

Sousa‐Lima

et al. 2022

Eur J Clin Investigation

Self Cite

View full text Add to dashboard Cite

show abstract

“…Current type 2 diabetes diagnosis criteria and therapeutic targets are mainly focused on glycemia and Hemoglobin A1c (HbA1c) measurements, but improvements in health outcomes demand tools that rationally stratify the different states of dysglycemia. In response to these needs, cluster analysis, a powerful data mining tool, allows exploration and rationalization of complex data, through stratification into groups, maximizing similarity within the group and minimizing it between groups [5].…”

Section: Introductionmentioning

confidence: 99%

Metabolic Footprint, towards Understanding Type 2 Diabetes beyond Glycemia

Pina

Patarrão

Ribeiro

et al. 2020

JCM

Self Cite

View full text Add to dashboard Cite

Type 2 diabetes (T2D) heterogeneity is a major determinant of complications risk and treatment response. Using cluster analysis, we aimed to stratify glycemia within metabolic multidimensionality and extract pathophysiological insights out of metabolic profiling. We performed a cluster analysis to stratify 974 subjects (PREVADIAB2 cohort) with normoglycemia, prediabetes, or non-treated diabetes. The algorithm was informed by age, anthropometry, and metabolic milieu (glucose, insulin, C-peptide, and free fatty acid (FFA) levels during the oral glucose tolerance test OGTT). For cluster profiling, we additionally used indexes of metabolism mechanisms (e.g., tissue-specific insulin resistance, insulin clearance, and insulin secretion), non-alcoholic fatty liver disease (NAFLD), and glomerular filtration rate (GFR). We found prominent heterogeneity within two optimal clusters, mainly representing normometabolism (Cluster-I) or insulin resistance and NAFLD (Cluster-II), at higher granularity. This was illustrated by sub-clusters showing similar NAFLD prevalence but differentiated by glycemia, FFA, and GFR (Cluster-II). Sub-clusters with similar glycemia and FFA showed dissimilar insulin clearance and secretion (Cluster-I). This work reveals that T2D heterogeneity can be captured by a thorough metabolic milieu and mechanisms profiling—metabolic footprint. It is expected that deeper phenotyping and increased pathophysiology knowledge will allow to identify subject’s multidimensional profile, predict their progression, and treat them towards precision medicine.

show abstract

The impact of hypoglycaemia on daily functioning among adults with diabetes: a prospective observational study using the Hypo-METRICS app

Søholm,

Broadley,

Zaremba

et al. 2024

Diabetologia

View full text Add to dashboard Cite

Aims/hypothesis The aim of this work was to examine the impact of hypoglycaemia on daily functioning among adults with type 1 diabetes or insulin-treated type 2 diabetes, using the novel Hypo-METRICS app. Methods For 70 consecutive days, 594 adults (type 1 diabetes, n=274; type 2 diabetes, n=320) completed brief morning and evening Hypo-METRICS ‘check-ins’ about their experienced hypoglycaemia and daily functioning. Participants wore a blinded glucose sensor (i.e. data unavailable to the participants) for the study duration. Days and nights with or without person-reported hypoglycaemia (PRH) and/or sensor-detected hypoglycaemia (SDH) were compared using multilevel regression models. Results Participants submitted a mean ± SD of 86.3±12.5% morning and 90.8±10.7% evening check-ins. For both types of diabetes, SDH alone had no significant associations with the changes in daily functioning scores. However, daytime and night-time PRH (with or without SDH) were significantly associated with worsening of energy levels, mood, cognitive functioning, negative affect and fear of hypoglycaemia later that day or while asleep. In addition, night-time PRH (with or without SDH) was significantly associated with worsening of sleep quality (type 1 and type 2 diabetes) and memory (type 2 diabetes). Further, daytime PRH (with or without SDH), was associated with worsening of fear of hyperglycaemia while asleep (type 1 diabetes), memory (type 1 and type 2 diabetes) and social functioning (type 2 diabetes). Conclusions/interpretation This prospective, real-world study reveals impact on several domains of daily functioning following PRH but not following SDH alone. These data suggest that the observed negative impact is mainly driven by subjective awareness of hypoglycaemia (i.e. PRH), through either symptoms or sensor alerts/readings and/or the need to take action to prevent or treat episodes. Graphical Abstract

show abstract

Clustering Clinical Data in R

Cited by 9 publications

References 46 publications

Big data and machine learning to tackle diabetes management

Big data and machine learning to tackle diabetes management

Metabolic Footprint, towards Understanding Type 2 Diabetes beyond Glycemia

The impact of hypoglycaemia on daily functioning among adults with diabetes: a prospective observational study using the Hypo-METRICS app

Contact Info

Product

Resources

About