Prediction of caregiver quality of life in amyotrophic lateral sclerosis using explainable machine learning

Artificial Intelligence Support for Informal Patient Caregivers: A Systematic Review

This study aims to explore how artificial intelligence can help ease the burden on caregivers, filling a gap in current research and healthcare practices due to the growing challenge of an aging population and increased reliance on informal caregivers. We conducted a search with Google Scholar, PubMed, Scopus, IEEE Xplore, and Web of Science, focusing on AI and caregiving. Our inclusion criteria were studies where AI supports informal caregivers, excluding those solely for data collection. Adhering to PRISMA 2020 guidelines, we eliminated duplicates and screened for relevance. From 947 initially identified articles, 10 met our criteria, focusing on AI's role in aiding informal caregivers. These studies, conducted between 2012 and 2023, were globally distributed, with 80% employing machine learning. Validation methods varied, with Hold-Out being the most frequent. Metrics across studies revealed accuracies ranging from 71.60% to 99.33%. Specific methods, like SCUT in conjunction with NNs and LibSVM, showcased accuracy between 93.42% and 95.36% as well as F-measures spanning 93.30% to 95.41%. AUC values indicated model performance variability, ranging from 0.50 to 0.85 in select models. Our review highlights AI's role in aiding informal caregivers, showing promising results despite different approaches. AI tools provide smart, adaptive support, improving caregivers' effectiveness and well-being.

Mediating effect of social support between caregiver burden and quality of life among family caregivers of cancer patients in palliative care units

PURPOSE

To identify factors influencing the quality of life of family caregivers with terminal cancer in Chinese palliative wards and to test whether social support mediates the relationship between caregiver burden and caregiver quality of life.

METHODS

A cross-sectional study design was used. Sociodemographic data were collected and the Quality of Life Scale, the Caregiver Burden Scale, and the Social Support Rating Scale were administered to Chinese family caregivers from December 2021 to December 2022. The factors influencing quality of life and caregiver burden were examined using the Mann‒Whitney U test and the Kruskal‒Wallis H test. The mediating role of social support was assessed using the bootstrap method.

RESULTS

Family caregivers' quality of life in Chinese terminal cancer palliative units was related to caregivers' daily care time, the caregiver-patient relationship, and patient age. Caregiver quality of life was negatively associated with caregiver burden and positively associated with social support. In addition, social support mediated the relationship between caregiver burden and caregiver quality of life.

CONCLUSION

Social support mediated the impact of caregiver burden on caregiver quality of life. Family, society, and palliative care institutions should be integrated to take actions to reduce family caregiver burden, increase social support, and transfer the positive aspects of specific cultural contexts to the culture of palliative care in general to collaboratively cope with various problems related to end-stage cancer.

Machine-learning model predicting quality of life using multifaceted lifestyles in middle-aged South Korean adults: a cross-sectional study

BACKGROUND

In the context of population aging, advances in healthcare technology, and growing interest in healthy aging and higher quality of life (QOL), have gained central focus in public health, particularly among middle-aged adults.

METHODS

This study presented an optimal prediction model for QOL among middle-aged South Korean adults (N = 4,048; aged 30-55 years) using a machine-learning technique. Community-based South Korean population data were sampled through multistage stratified cluster sampling. Twenty-one variables related to individual factors and various lifestyle patterns were surveyed. QOL was assessed using the Short Form Health Survey (SF-12) and categorized into total QOL, physical component score (PCS), and mental component score (MCS). Seven machine-learning algorithms were used to predict QOL: decision tree, Gaussian Naïve Bayes, k-nearest neighbor, logistic regression, extreme gradient boosting, random forest, and support vector machine. Data imbalance was resolved with the synthetic minority oversampling technique (SMOTE). Random forest was used to compare feature importance and visualize the importance of each variable.

RESULTS

For predicting QOL deterioration, the random forest method showed the highest performance. The random forest algorithm using SMOTE showed the highest area under the receiver operating characteristic (AUC) for total QOL (0.822), PCS (0.770), and MCS (0.786). Applying the data, SMOTE enhanced model performance by up to 0.111 AUC. Although feature importance differed across the three QOL indices, stress and sleep quality were identified as the most potent predictors of QOL. Random forest generated the most accurate prediction of QOL among middle-aged adults; the model showed that stress and sleep quality management were essential for improving QOL.

CONCLUSION

The results highlighted the need to develop a health management program for middle-aged adults that enables multidisciplinary management of QOL.

Derivation and validation of risk prediction for posttraumatic stress symptoms following trauma exposure

BACKGROUND

Posttraumatic stress symptoms (PTSS) are common following traumatic stress exposure (TSE). Identification of individuals with PTSS risk in the early aftermath of TSE is important to enable targeted administration of preventive interventions. In this study, we used baseline survey data from two prospective cohort studies to identify the most influential predictors of substantial PTSS.

METHODS

Self-identifying black and white American women and men (n = 1546) presenting to one of 16 emergency departments (EDs) within 24 h of motor vehicle collision (MVC) TSE were enrolled. Individuals with substantial PTSS (⩾33, Impact of Events Scale - Revised) 6 months after MVC were identified via follow-up questionnaire. Sociodemographic, pain, general health, event, and psychological/cognitive characteristics were collected in the ED and used in prediction modeling. Ensemble learning methods and Monte Carlo cross-validation were used for feature selection and to determine prediction accuracy. External validation was performed on a hold-out sample (30% of total sample).

RESULTS

Twenty-five percent (n = 394) of individuals reported PTSS 6 months following MVC. Regularized linear regression was the top performing learning method. The top 30 factors together showed good reliability in predicting PTSS in the external sample (Area under the curve = 0.79 ± 0.002). Top predictors included acute pain severity, recovery expectations, socioeconomic status, self-reported race, and psychological symptoms.

CONCLUSIONS

These analyses add to a growing literature indicating that influential predictors of PTSS can be identified and risk for future PTSS estimated from characteristics easily available/assessable at the time of ED presentation following TSE.

Application of explainable artificial intelligence for healthcare: A systematic review of the last decade (2011-2022)

BACKGROUND AND OBJECTIVES

Artificial intelligence (AI) has branched out to various applications in healthcare, such as health services management, predictive medicine, clinical decision-making, and patient data and diagnostics. Although AI models have achieved human-like performance, their use is still limited because they are seen as a black box. This lack of trust remains the main reason for their low use in practice, especially in healthcare. Hence, explainable artificial intelligence (XAI) has been introduced as a technique that can provide confidence in the model's prediction by explaining how the prediction is derived, thereby encouraging the use of AI systems in healthcare. The primary goal of this review is to provide areas of healthcare that require more attention from the XAI research community.

METHODS

Multiple journal databases were thoroughly searched using PRISMA guidelines 2020. Studies that do not appear in Q1 journals, which are highly credible, were excluded.

RESULTS

In this review, we surveyed 99 Q1 articles covering the following XAI techniques: SHAP, LIME, GradCAM, LRP, Fuzzy classifier, EBM, CBR, rule-based systems, and others.

CONCLUSION

We discovered that detecting abnormalities in 1D biosignals and identifying key text in clinical notes are areas that require more attention from the XAI research community. We hope this is review will encourage the development of a holistic cloud system for a smart city.

Explainable AI for clinical and remote health applications: a survey on tabular and time series data

Nowadays Artificial Intelligence (AI) has become a fundamental component of healthcare applications, both clinical and remote, but the best performing AI systems are often too complex to be self-explaining. Explainable AI (XAI) techniques are defined to unveil the reasoning behind the system’s predictions and decisions, and they become even more critical when dealing with sensitive and personal health data. It is worth noting that XAI has not gathered the same attention across different research areas and data types, especially in healthcare. In particular, many clinical and remote health applications are based on tabular and time series data, respectively, and XAI is not commonly analysed on these data types, while computer vision and Natural Language Processing (NLP) are the reference applications. To provide an overview of XAI methods that are most suitable for tabular and time series data in the healthcare domain, this paper provides a review of the literature in the last 5 years, illustrating the type of generated explanations and the efforts provided to evaluate their relevance and quality. Specifically, we identify clinical validation, consistency assessment, objective and standardised quality evaluation, and human-centered quality assessment as key features to ensure effective explanations for the end users. Finally, we highlight the main research challenges in the field as well as the limitations of existing XAI methods.

"Walking a tightrope": A grounded theory approach to informal caregiving for amyotrophic lateral sclerosis

Informal caregivers, mainly family members and friends, provide supportive and palliative care for people with amyotrophic lateral sclerosis (ALS) during their terminal disease course. Informal caregiving for people with ALS continues towards palliative care and end-of-life care with the progression of the disease. In this study, we provide a theoretical understanding of informal caregiving in ALS utilising 23 semi-structured interviews conducted with informal caregivers of people with ALS (pwALS) in Switzerland. Due to the expected death of the care recipient, our grounded theory approach outlines informal caregivers' caregiving work as an effort to secure a balance amongst different caregiving activities, which feed into the final stage of providing palliative care. Overall, our theoretical understanding of ALS informal caregiving work encompasses the core category 'holding the balance' and four secondary categories: 'Organising support', 'being present', 'managing everyday life' and 'keeping up with ALS'. The core category of holding the balance underlines the significance of ensuring care and normalcy even as disease progresses and until the end of life. For the informal caregivers, this balancing act is the key element of care provision to pwALS and therefore guides decisions surrounding caregiving. On this understanding, those caregivers that succeed in holding the balance can provide care at home until death. The balance is heavily influenced by contextual factors of caregiving, for example relating to personal characteristics of the caregiver, or activities of caregiving where the goal is to ensure the quality of life of the pwALS. As there is a heterogeneity of speed and subtype of progression of ALS, our work accounts for multiple caregiving trajectories.

Health-Related Quality of Life and Perceived Burden of Informal Caregivers of Patients with Rare Diseases in Selected European Countries

Most of rare disease (RD) patients are assisted in their homes by their family as informal caregivers, causing a substantial burden among family members devoted to care. The role of informal caregivers has been associated with increased levels of stress, poor physical/mental health and impaired HRQOL. The present study assessed the impact on HRQOL and perceived burden of long-term informal caregiving, as well as the inter-relationships of individuals affected by different RD in six European countries, taking advantage of the data provided by the BURQOL-RD project (France, Germany, Italy, Spain, Sweden and UK). Correlation analysis was used to explore the relation between caregiver HRQOL and caregiver burden (Zarit Burden Interview). Multinomial logistic regression models were used to explore the role of explanatory variables on each domain of caregivers HRQOL measured by EQ-5D. Caregivers' HRQOL is inversely correlated with burden of caring. Mobility dimension of EQ-5D was significantly associated with patients age, time devoted to care by secondary caregivers, patient gender and patient utility index. Patients' age, burden scores and patient utility index significantly predict the capacity of caregivers to perform activities of daily living. Employed caregivers are less likely of reporting 'slight problems' in pain/discomfort dimensions than unemployed caregivers. The EQ-5D instrument is sensitive to measure differences in HRQOL between caregivers with different levels of burden of care.

A Clinical Decision Support System for the Prediction of Quality of Life in ALS

Amyotrophic Lateral Sclerosis (ALS), also known as Motor Neuron Disease (MND), is a rare and fatal neurodegenerative disease. As ALS is currently incurable, the aim of the treatment is mainly to alleviate symptoms and improve quality of life (QoL). We designed a prototype Clinical Decision Support System (CDSS) to alert clinicians when a person with ALS is experiencing low QoL in order to inform and personalise the support they receive. Explainability is important for the success of a CDSS and its acceptance by healthcare professionals. The aim of this work isto announce our prototype (C-ALS), supported by a first short evaluation of its explainability. Given the lack of similar studies and systems, this work is a valid proof-of-concept that will lead to future work. We developed a CDSS that was evaluated by members of the team of healthcare professionals that provide care to people with ALS in the ALS/MND Multidisciplinary Clinic in Dublin, Ireland. We conducted a user study where participants were asked to review the CDSS and complete a short survey with a focus on explainability. Healthcare professionals demonstrated some uncertainty in understanding the system’s output. Based on their feedback, we altered the explanation provided in the updated version of our CDSS. C-ALS provides local explanations of its predictions in a post-hoc manner, using SHAP (SHapley Additive exPlanations). The CDSS predicts the risk of low QoL in the form of a probability, a bar plot shows the feature importance for the specific prediction, along with some verbal guidelines on how to interpret the results. Additionally, we provide the option of a global explanation of the system’s function in the form of a bar plot showing the average importance of each feature. C-ALS is available online for academic use.

An explainable machine learning-based clinical decision support system for prediction of gestational diabetes mellitus

Gestational Diabetes Mellitus (GDM), a common pregnancy complication associated with many maternal and neonatal consequences, is increased in mothers with overweight and obesity. Interventions initiated early in pregnancy can reduce the rate of GDM in these women, however, untargeted interventions can be costly and time-consuming. We have developed an explainable machine learning-based clinical decision support system (CDSS) to identify at-risk women in need of targeted pregnancy intervention. Maternal characteristics and blood biomarkers at baseline from the PEARS study were used. After appropriate data preparation, synthetic minority oversampling technique and feature selection, five machine learning algorithms were applied with five-fold cross-validated grid search optimising the balanced accuracy. Our models were explained with Shapley additive explanations to increase the trustworthiness and acceptability of the system. We developed multiple models for different use cases: theoretical (AUC-PR 0.485, AUC-ROC 0.792), GDM screening during a normal antenatal visit (AUC-PR 0.208, AUC-ROC 0.659), and remote GDM risk assessment (AUC-PR 0.199, AUC-ROC 0.656). Our models have been implemented as a web server that is publicly available for academic use. Our explainable CDSS demonstrates the potential to assist clinicians in screening at risk patients who may benefit from early pregnancy GDM prevention strategies.

Psychological Support for Family Caregivers of Patients With Amyotrophic Lateral Sclerosis at the Time of the Coronavirus Disease 2019 Pandemic: A Pilot Study Using a Telemedicine Approach

The coronavirus disease 2019 (COVID-19) pandemic confined most of the population to homes worldwide, and then, a lot of amyotrophic lateral sclerosis (ALS) centers moved to telemedicine services to continue to assist both patients with ALS and their caregivers. This pilot, randomized, controlled study aimed to explore the potential role of psychological support interventions for family caregivers of patients with ALS through resilience-oriented sessions of group therapy during the COVID-19 pandemic. In total, 12 caregivers agreed to be remotely monitored by our center since March 2020 and underwent scales for global burden (i.e., Caregiver Burden Inventory, CBI), resilience (i.e., Connor Davidson Resilience Scale, CD-RISC), and perceived stress (i.e., Perceived Stress Scale, PSS) at two-time points (i.e., at pre-treatment assessment and after 9 months or at post-treatment assessment). They were randomized into two groups: the former group underwent resilience-oriented sessions of group therapy two times a month for 3 months, while the latter one was only remotely monitored. No significant differences were found in CBI, CD-RISC, and PSS during the 9-month observation period in the treated group compared with the control group, suggesting a trend toward stability of caregiver burden together with resilience and perceived stress scores in all the subjects monitored. The lack of differences in caregivers' burden, resilience, and perceived stress scores by comparing the two groups monitored during 9 months could be due to the co-occurrence of the COVID-19 pandemic with the stressful events related to caring for patients with ALS that might have hindered the detection of significant benefits from short-lasting psychological support.