Lauri Lahti

Lauri Lahti

We propose and experimentally motivate a new methodology to support decision-making processes in healthcare with artificial intelligence based on personal rankings of care decision making steps that can be identified with our methodology, questionnaire data and its statistical patterns. Our longitudinal quantitative cross-sectional three-stage study gathered self-ratings for 437 expression statements concerning healthcare situations on Likert scales in respect to "the need for help", "the advancement of health", "the hopefulness", "the indication of compassion" and "the health condition", and 45 answers about the person's demographics, health and wellbeing, also the duration of giving answers. Online respondents between 1 June 2020 and 29 June 2021 were recruited from Finnish patient and disabled people's organizations, other health-related organizations and professionals, and educational institutions (n=1075). With Kruskal-Wallis test, Wilcoxon rank-sum test (i.e., Mann-Whitney U test), Wilcoxon rank-sum pairwise test, Welch's t test and one-way analysis of variance (ANOVA) between groups test we identified statistically significant differences of ratings and their durations for each expression statement in respect to respondent groupings based on the answer values of each background question. Frequencies of the later reordering of rating rankings showed dependencies with ratings given earlier in respect to various interpretation task entities, interpretation dimensions and respondent groupings. Our methodology, questionnaire data and its statistical patterns enable analyzing with self-rated expression statements the representations of decision making steps in healthcare situations and their chaining, agglomeration and branching in knowledge entities of personalized care paths. Our results support building artificial intelligence solutions to address the patient's needs concerning care.

Lauri Lahti

Developing machine learning models to support health analytics requires increased understanding about statistical properties of self-rated expression statements. We analyzed self-rated expression statements concerning the coronavirus COVID-19 epidemic to identify statistically significant differences between groups of respondents and to detect the patient's need for help with machine learning. Our quantitative study gathered the "need for help" ratings for twenty health-related expression statements concerning the coronavirus epidemic on a 11-point Likert scale, and nine answers about the person's health and wellbeing, sex and age. Online respondents between 30 May and 3 August 2020 were recruited from Finnish patient and disabled people's organizations, other health-related organizations and professionals, and educational institutions (n=673). We analyzed rating differences and dependencies with Kendall rank-correlation and cosine similarity measures and tests of Wilcoxon rank-sum, Kruskal-Wallis and one-way analysis of variance (ANOVA) between groups, and carried out machine learning experiments with a basic implementation of a convolutional neural network algorithm. We found statistically significant correlations and high cosine similarity values between various health-related expression statement pairs concerning the "need for help" ratings and a background question pair. We also identified statistically significant rating differences for several health-related expression statements in respect to groupings based on the answer values of background questions, such as the ratings of suspecting to have the coronavirus infection and having it depending on the estimated health condition, quality of life and sex. Our experiments with a convolutional neural network algorithm showed the applicability of machine learning to support detecting the need for help in the patient's expressions.