Predicting dementia diagnosis from cognitive footprints in electronic health records: a case-control study protocol

Comparisons of the risks of new-onset prostate cancer in type 2 diabetes mellitus between SGLT2I and DPP4I users: A population-based cohort study

BACKGROUND

Sodium-glucose cotransporter 2 inhibitors (SGLT2I) have been suggested to reduce new-onset cancer amongst type-2 diabetes mellitus (T2DM) patients. This study aims to compare the risks of prostate cancer between SGLT2I and dipeptidyl peptidase-4 inhibitors (DPP4I) amongst T2DM patients.

DESIGN, SETTING AND PARTICIPANTS

This was a retrospective population-based cohort study of prospectively recorded data on male patients with T2DM who were prescribed either SGLT2I or DPP4I between 1st January 2015 and 31st December 2020 from Hong Kong.

METHODS

The primary outcome was new-onset prostate cancer. The secondary outcomes included cancer-related mortality and all-cause mortality. Propensity score matching (1:1 ratio) using the nearest neighbor search was performed and multivariable Cox regression was applied. A three-arm analysis including the glucagon-like peptide-1 receptor agonist (GLP1a) cohort was conducted.

RESULTS

This study included 42129 male T2DM patients (median age: 61.0 years old [SD: 12.2]; SGLT2I: n = 17,120; DPP4I: n = 25,009). In the propensity score matched cohort, the number of prostate cancers was significantly lower in SGLT2I users (n = 60) than in DPP4I (n = 102). Over a follow-up duration of 5.61 years, SGLT2I was associated with lower prostate cancer risks (HR: 0.45; 95% CI: 0.30-0.70) than DPP4I after adjustments. The subgroup analyses showed that the interactions between SGLT2I and age, hypertension, heart failure, and GLP-1a were not statistically significant. The result remained consistent in the sensitivity analysis.

CONCLUSION

The study demonstrated SGLT2I was associated with lower risks of new-onset prostate cancer after propensity score matching and adjustments compared to DPP4I amongst T2DM patients.

Predicting Alzheimer's disease from cognitive footprints in mid and late life: How much can register data and machine learning help?

BACKGROUND

Real-world data with decades-long medical records are increasingly available alongside the growing adoption of machine learning in healthcare research. We evaluated the performance of machine learning models in predicting the risk of Alzheimer's disease (AD) using data from the Finnish national registers.

METHODS

We conducted a case-control study using data from the Finnish MEDALZ (Medication use and Alzheimer's disease) study. Altogether 56,741 individuals with incident AD diagnosis (age ≥ 65 years at diagnosis and born after 1922) and their 1:1 age-, sex-, and region of residence-matched controls were included. The association of risk factors, evaluated at different age periods (45-54, 55-64, 65+), and AD were assessed with logistic regression. Predictive accuracies of logistic regressions were compared with seven machine learning models (L1-regularized logistic regression, Naive bayes, Decision tree, Random Forest, Multilayer perceptron, XGBoost, and LightGBM).

FINDINGS

63.5 % of cases and controls were females and the mean age was 79.1 (SD = 5.1). The strongest associations with AD were observed for head injuries at age 55-64 (OR, 95 % CI 1.33, 1.19-1.48) and 65+ (1.31, 1.23-1.40), followed by antidepressant use (1.30, 1.22-1.38) at 55-64 and antipsychotic use (1.27, 1.19-1.35) at 65+. The predictive accuracies of all models were low, with the best performance (AUC 0.603) observed in Random Forest for predicting AD onset at age 65-69.

INTERPRETATION

Although significant associations were identified between many risk factors and AD, the low predictive accuracies suggest that specialised healthcare diagnosis data is not sufficient for predicting AD and linkage with other data sources is needed.

Identifying dementia from cognitive footprints in hospital records among Chinese older adults: a machine-learning study

Background

By combining theory-driven and data-driven methods, this study aimed to develop dementia predictive algorithms among Chinese older adults guided by the cognitive footprint theory.

Methods

Electronic medical records from the Clinical Data Analysis and Reporting System in Hong Kong were employed. We included patients with dementia diagnosed at 65+ between 2010 and 2018, and 1:1 matched dementia-free controls. We identified 51 features, comprising exposures to established modifiable factors and other factors before and after 65 years old. The performances of four machine learning models, including LASSO, Multilayer perceptron (MLP), XGBoost, and LightGBM, were compared with logistic regression models, for all patients and subgroups by age.

Findings

A total of 159,920 individuals (40.5% male; mean age [SD]: 83.97 [7.38]) were included. Compared with the model included established modifiable factors only (area under the curve [AUC] 0.689, 95% CI [0.684, 0.694]), the predictive accuracy substantially improved for models with all factors (0.774, [0.770, 0.778]). Machine learning and logistic regression models performed similarly, with AUC ranged between 0.773 (0.768, 0.777) for LASSO and 0.780 (0.776, 0.784) for MLP. Antipsychotics, education, antidepressants, head injury, and stroke were identified as the most important predictors in the total sample. Age-specific models identified different important features, with cardiovascular and infectious diseases becoming prominent in older ages.

Interpretation

The models showed satisfactory performances in identifying dementia. These algorithms can be used in clinical practice to assist decision making and allow timely interventions cost-effectively.

Funding

The Research Grants Council of Hong Kong under the Early Career Scheme 27110519.

Discrimination of the Cognitive Function of Community Subjects Using the Arterial Pulse Spectrum and Machine-Learning Analysis

Early identification of cognitive impairment would allow affected patients to receive care at earlier stage. Changes in the arterial stiffness have been identified as a prominent pathological feature of dementia. This study aimed to verify if applying machine-learning analysis to spectral indices of the arterial pulse waveform can be used to discriminate different cognitive conditions of community subjects. 3-min Radial arterial blood pressure waveform (BPW) signals were measured noninvasively in 123 subjects. Eight machine-learning algorithms were used to evaluate the following 4 pulse indices for 10 harmonics (total 40 BPW spectral indices): amplitude proportion and its coefficient of variation; phase angle and its standard deviation. Significant differences were noted in the spectral pulse indices between Alzheimer’s-disease patients and control subjects. Using them as training data (AUC = 70.32% by threefold cross-validation), a significant correlation (R² = 0.36) was found between the prediction probability of the test data (comprising community subjects at two sites) and the Mini-Mental-State-Examination score. This finding illustrates possible physiological connection between arterial pulse transmission and cognitive function. The present findings from pulse-wave and machine-learning analyses may be useful for discriminating cognitive condition, and hence in the development of a user-friendly, noninvasive, and rapid method for the early screening of dementia.

Accuracy of Support-Vector Machines for Diagnosis of Alzheimer's Disease, Using Volume of Brain Obtained by Structural MRI at Siriraj Hospital

Background: The determination of brain volumes using visual ratings is associated with an inherently low accuracy for the diagnosis of Alzheimer's disease (AD). A support-vector machine (SVM) is one of the machine learning techniques, which may be utilized as a classifier for various classification problems. This study exploratorily investigated the accuracy of SVM classification models for AD subjects using brain volume and various clinical data as features.

Methods: The study was designed as a retrospective chart review. A total of 201 eligible subjects were recruited from the Memory Clinic at Siriraj Hospital, Thailand. Eighteen cases were excluded due to incomplete MRI data. Subjects were randomly assigned to a training group (AD = 46, normal = 46) and testing group (AD = 45, normal = 46) for SVM modeling and validation, respectively. The results in terms of accuracy and a receiver operating characteristic curve analysis are reported.

Results: The highest accuracy for brain volumetry (62.64%) was found using the hippocampus as a single feature. A combination of clinical parameters as features provided accuracy ranging between 83 and 90%. However, a combination of brain volumetry and clinical parameters as features to the SVM models did not improve the accuracy of the result.

Conclusions: In our study, the use of brain volumetry as SVM features provided low classification accuracy with the highest accuracy of 62.64% using the hippocampus volume alone. In contrast, the use of clinical parameters [Thai mental state examination score, controlled oral word association tests (animals; and letters K, S, and P), learning memory, clock-drawing test, and construction-praxis] as features for SVM models provided good accuracy between 83 and 90%.

The Landmark Series: Minimally Invasive Pancreatic Resection

BACKGROUND

Pancreatic resections are among the most technically demanding procedures, including a high risk of potentially life-threatening complications and outcomes strongly correlated to hospital volume and individual surgeon experience. Minimally invasive pancreatic resections (MIPRs) have become a part of standard surgical practice worldwide over the last decade; however, in comparison with other surgical procedures, the implementation of minimally invasive approaches into clinical practice has been rather slow.

OBJECTIVE

The aim of this study was to highlight and summarize the available randomized controlled trials (RCTs) evaluating the role of minimally invasive approaches in pancreatic surgery.

METHODS

A WHO trial registry and Pubmed database literature search was performed to identify all RCTs comparing MIPRs (robot-assisted and/or laparoscopic distal pancreatectomy [DP] or pancreatoduodenectomy [PD]) with open pancreatic resections (OPRs).

RESULTS

Overall, five RCTs on MIPR versus OPR have been published and seven RCTs are currently recruiting. For DP, the results of two RCTs were in favor of minimally invasive distal pancreatectomy (MIDP) in terms of shorter hospital stay and less intraoperative blood loss, with comparable morbidity and mortality. Regarding PD, two RCTs showed similar advantages for MIPD. However, concerns were raised after the early termination of the third multicenter RCT on MIPD versus open PD due to higher complication-related mortality in the laparoscopic group and no clear other demonstrable advantages. No RCTs on robot-assisted pancreatic procedures are available as yet.

CONCLUSION

At the current level of evidence, MIDP is thought to be safe and feasible, although oncological safety should be further evaluated. Based on the results of the RCTs conducted for PD, MIPD cannot be proclaimed as the superior alternative to open PD, although promising outcomes have been demonstrated by experienced centers. Future studies should provide answers to the role of robotic approaches in pancreatic surgery and aim to identity the subgroups of patients or indications with the greatest benefit of MIPRs.