MNMaisa Niemelä

Maisa Niemelä

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Maisa Niemelä has a total of 103 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Severity of depression and anxiety symptoms is reflected in physiological and behavioral metrics collected from a consumer-grade wearable ring

    by Saeid Azadifar, Alireza Sameh, Laura Nauha, Mikko Kärmeniemi, Maisa Niemelä, Vahid Farrahi

  • Open Access

    Wearable Prompt: In-Context Learning for Depression and Anxiety Prediction from Consumer Smart Ring Metrics

    by Saeid Azadifar, Alireza Sameh, Maisa Niemela, Vahid Farrahi

    Large language models provide a promising framework for wearable-based health prediction by converting structured physiological and behavioral measurements into natural-language prompts. In this paper, we investigate whether pre-trained lightweight open-weight LLMs can predict depression and anxiety symptoms from short-horizon consumer wearable data. Using 4-8 days of Oura Ring data from 1,285 participants in the Northern Finland Birth Cohort 1986, we convert activity, sleep, heart rate, heart rate variability, demographic, and anthropometric measurements into structured prompts. We evaluate Llama 3.1, BioMistral, and Qwen 2.5 under zero-shot, rule-based, and few-shot in-context learning settings. To contextualize LLM performance, we compare them against machine learning models and recurrent neural networks. Our results show that prompt design is critical for LLM-based wearable inference. Zero-shot LLMs achieve high accuracy but largely predict the majority class, failing to identify participants with depression and anxiety symptoms. In contrast, few-shot prompting substantially improves positiveclass detection. Llama 3.1 with four in-context examples achieves the strongest performance, with 0.92 accuracy, 0.82 macro-F1, and 0.69 F1 for the positive class, among evaluated models. These findings suggest that lightweight LLMs can use in-context examples to better interpret structured wearable summaries and possibly provide a scalable direction for mental health prediction from consumer wearable data in combination with pre-trained LLMs.

  • Open Access

    Differences in domain-specific composition of 24-hour physical behaviors across occupational groups

    by Niko-Petteri Hirvonen, Maisa Niemelä, Anna-Maiju Leinonen, Vahid Farrahi

    Purpose: Physical behaviors, including physical activity, sedentary time, and sleep, are associated with health and together comprise the 24-hour day. Among working-aged adults, the composition of these 24-hour physical behaviors may vary across occupations, but the extent of these differences remains unknown. This population-based study examined how the composition of 24-hour physical behaviors differs in working-aged adults across occupational groups during work and leisure time. Methods: The study population consisted of 2,809 adults from the Northern Finland Birth Cohort 1966 46-year follow-up study, who wore a hip-worn accelerometer for 14 days. The accelerometry data was combined with self-reported diaries to determine time in bed and work hours. Four occupational groups were aggregated based on occupation type and occupational skill level: high-skill white-collar, low-skill white-collar, high-skill blue-collar and low-skill blue-collar. Time spent on physical behaviors during workdays (including working hours and leisure time separately) and non-workdays were calculated. Differences in physical behavior compositions were assessed using a compositional data analysis approach based on log-ratios of geometric means, with the low-skill blue-collar group as the reference group. Results: Significant differences in 24-hour physical behaviors across occupational groups were mainly observed during workdays. During workday working hours, all other occupational groups had less physical activity (both moderate-to-vigorous and light-intensity physical activity) compared to the reference group. During workday leisure time, significant differences were observed in all physical behaviors, except for sedentary time. On non-working days, the high-skill white collar group had higher moderate-to-vigorous physical activity compared to the reference group. Conclusions: The composition of domain-specific 24-hour physical behaviors differs across occupational groups, particularly during workdays.

  • Open Access

    Digital phenotyping using wearable-determined physical behaviours and machine learning to detect depression and anxiety symptoms in a general population

    by Alireza Sameh, Laura Nauha, Marjo Seppänen, Maisa Niemelä, Mourad Oussalah, Raija Korpelainen, Vahid Farrahi


  • Open Access

    Association of childhood residential environment and family socioeconomic status with physical activity in early adulthood - a population-based Northern Finland Birth Cohort 1986 study

    by Ulla-Maija Luoma, Soile Puhakka, Pyky Riitta, Anna-Maiju Leinonen, Maisa Niemelä, Raija Korpelainen, Tiina Lankila

  • Barrier Self-Efficacy Moderates the Relationship Between Self-Compassion and Self-Reported Leisure-Time Physical Activity

    by Sohvi Nuojua, Anna-Maiju Leinonen, Raija Korpelainen, Maisa Niemelä, Marjo Seppänen, Mirka Hintsanen

    Self-compassion, a compassionate attitude toward oneself, has been associated with positive physical activity (PA) outcomes. This cross-sectional study on the 33- to 35-year follow-up data of the Northern Finland Birth Cohort 1986 examined the associations between self-compassion and self-reported leisure-time PA (LTPA), accelerometer-measured PA, and self-reported sitting time and assessed the moderating effect of PA barrier self-efficacy on these associations. Self-compassion was associated with higher LTPA ( N = 2,080), also after adjusting for key PA-related covariates, such as gender, perceived health, and level of education but not after adjusting for barrier self-efficacy. Self-compassion was also associated with higher accelerometer-measured PA ( N = 811), and lower sitting time ( N = 1,072), but not after adjusting for covariates. Barrier self-efficacy moderated the associations between self-compassion and self-reported activity behaviors, but not accelerometer-measured PA. Higher self-compassion was associated with higher LTPA and lower sitting time in individuals with high barrier self-efficacy. Conversely, higher self-compassion signified lower LTPA in individuals with low barrier self-efficacy. Accounting for barrier self-efficacy is thus necessary for the beneficial role of self-compassion in PA to be realized.

  • Open Access

    Severity of Depression and Anxiety Symptoms Manifest in Physiological and Behavioral Metrics Collected from a Consumer-Grade Wearable Ring

    by Alireza Sameh, Saeid Azadifar, Laura Nauha, Mikko Kärmeniemi, Maisa Niemelä, Vahid Farrahi

    Wearable devices can collect changes in human behaviors related to mental health including depression and anxiety. Here, we examined whether and how digital metrics from a consumer-grade wearable smart ring (Oura Ring) differed by severity of depression and anxiety symptoms using data from a large-scale population-based sample of young adults (n=1,290, age range: 33–35). Participants wore the ring for two weeks, assessing sleep architecture, nocturnal heart rate (HR), heart rate variability (HRV), and movement intensity. Mental health symptoms were assessed using the Generalized Anxiety Disorder 7-item and Hopkins Symptom Checklist-25 scales. On average, participants with higher depression and/or anxiety symptoms had lower levels of rapid eye movement and had higher levels of deep and light sleep, elevated nocturnal HR, reduced HRV, and lower daytime movement compared to non-symptom individuals. Findings suggest that symptoms of depression and anxiety may manifest in physiological and behavioral metrics collected by consumer-grade wearable devices.

  • Open Access

    ASSOCIATION BETWEEN SLEEP REGULARITY AND SYMPTOMS OF ANXIETY AND DEPRESSION IN MIDDLE-AGED ADULTS

    by Samir Chalise, Laura Nauha, Raija Korpelainen, Maisa Niemela, Vahid Farrahi

    Background Sleep regularity plays a significant role in mental health, but its association with anxiety or depression symptoms in midlife and whether movement behaviors moderate these associations is unclear. We examined how variability in sleep timing and duration is associated with these symptoms and tested the moderating effects of physical activity and sedentary time. Methods We analyzed data from 3,556 46-year-old participants in the Northern Finland Birth Cohort 1966. Sleep regularity was derived from 7 days of accelerometer‑measured sleep timing (midpoint of sleep period) and sleep duration, expressed as standard deviations and categorized as regular, moderately irregular, or highly irregular. Anxiety and depression symptoms were assessed using the Hopkins Symptom Checklist‑25. Associations were examined using multivariable linear regression adjusted for gender, marital status, education, smoking, alcohol use, medication, chronotype, work schedule, and physical activity or sedentary time derived from the same accelerometer data, with moderation tested using interaction terms (B coefficients with 95% confidence intervals). Results Greater sleep timing irregularity was associated with higher anxiety symptoms (B = 0.025, 95% CI [0.007, 0.043], p = 0.001), and with depressive symptoms only when models included sedentary time (B = 0.035, 95% CI [0.017, 0.053], p < 0.001). Sleep duration irregularity showed no significant associations. Physical activity and sedentary time did not moderate these relationships. Conclusions Sleep timing regularity was more consistently associated with mental health outcomes, especially anxiety, than sleep duration variability. Maintaining regular sleep timing may support anxiety management in midlife, highlighting its potential relevance for preventive intervention.

  • Open Access

    Compositional Associations of 24‐h Movement Behaviors With Depressive and Anxiety Symptoms in Middle‐Aged Adults

    by Clarence Tan, Maisa Niemelä, Marjo Seppänen, Anna-Maiju Leinonen, Vahid Farrahi

    Background: Physical activity could reduce the risk of depression and anxiety. A 24‐h day includes time spent sleeping, in sedentary behavior (SB), or engaging in physical activities. However, the joint and combined associations of these 24‐h movement behaviors with depressive and anxiety symptoms remain unclear. Aim: To investigate the compositional associations of 24‐h movement behaviors with depressive and anxiety symptoms in a population‐based sample of middle‐aged adults. Methods: The study population ( N = 4490) comprised of participants from the Northern Finland Birth Cohort 1966 (NFBC1966). Over 14 consecutive days, movement behaviors—SB, light physical activity (LPA), and moderate‐to‐vigorous physical activity (MVPA)—were recorded using a hip‐worn accelerometer and were combined with self‐reported sleep duration to obtain the 24‐h time‐use composition. Three different screening tools—Beck Depression Inventory‐II (BDI‐II), Generalized Anxiety Disorder‐7 (GAD‐7), and Hopkins Symptom Checklist‐25 (HSCL‐25)—were utilized to assess the severity of depressive and anxiety symptoms. Compositional linear regression and time reallocation analysis were performed to examine the associations of 24‐h movement behaviors with depressive and anxiety symptoms. Results: Compositional 24‐h movement behaviors were significantly associated with depressive and anxiety symptoms. Higher MVPA relative to other components of 24‐h movement behaviors was associated with significantly lower depressive and anxiety symptoms. A 30‐min reallocation of time from SB to MVPA was associated with 9.0% (95% confidence interval [CI]: 6.3%, 11.6%) lower depressive symptoms and 4.5% (95% CI: 2.1%, 6.9%) lower anxiety symptoms assessed with BDI‐II and GAD‐7 tools, respectively. Similar patterns of associations were observed in the time reallocation analysis when depressive and anxiety symptoms were assessed with the HSCL‐25 screening tool. Conclusion: Each component of 24‐h movement behaviors was associated with depressive and anxiety symptoms. Among the component parts of 24‐h movement behaviors, MVPA was most strongly associated with lower depressive and anxiety symptoms, whereas LPA only showed marginal favorable benefits.


  • Open Access

    Dispositional optimism and pessimism in relation to physical activity and fitness in adulthood – a 15-year follow-up study

    by Jutta Karhu, Juha Veijola, Anna-Maiju Leinonen, Maisa Niemelä, Tuija H. Tammelin, Mirka Hintsanen

    Dispositional optimism and pessimism have been linked to physical activity (PA) in adulthood, but longitudinal associations and relationships with physical fitness (PF) remain underexplored. We investigated whether optimism and pessimism are a) longitudinally associated with self-reported leisure-time physical activity (LTPA), b) cross-sectionally associated with device-measured PA and c) cross-sectionally and longitudinally associated with clinically measured PF. Additionally, potential bidirectional associations were explored. A 15-year follow-up data (from age 31 to 46) was analysed from the Northern Finland Birth Cohort 1966 study (N = 7,146). Optimism and pessimism were assessed with the Life Orientation Test-Revised. Self-reported LTPA was measured with questionnaires at both time points. Device-measured PA was assessed with accelerometers at age 46. PF was measured clinically by cardiorespiratory fitness, hand-grip strength and trunk muscle endurance at both time points. Higher optimism and lower pessimism predicted higher volumes of LTPA over 15 years. Additionally, higher LTPA predicted lower pessimism. Optimism, but not pessimism, was cross-sectionally associated with device-measured PA at age 46. Regarding PF, cross-sectional associations were observed with handgrip strength and trunk muscle endurance. Higher optimism and lower pessimism promote PA in adulthood, while PA may reduce pessimism. Cross-sectional associations with muscular fitness were detected.