Longitudinal Dataset Research Articles

Swimming With the Shark: The Effects of Platform Price Promotion and In-Platform Advertising on Third-Party Retailer Performance in Hybrid Online Retailing

Hybrid retailing is an emerging business model that melds traditional reselling with the newer platform approach. A platform owner resells directly to consumers (the reselling model) and also allows third-party retailers (TPRs) to sell on the platform for a commission (the platform model). Given the integral role of TPRs in the hybrid retailing ecosystem, research is imperative to understand how the strategic actions of the platform owner could affect the performance and activities of TPRs. Using a longitudinal dataset of an online platform that adopts a hybrid-retailing structure, the research proposes that the platform owner's strategies, specifically price promotion and in-platform advertising, can affect TPR sales performance directly (through positive spillover effects and negative crowding-out effects) and indirectly (by prompting TPRs to respond with their price promotion and in-platform advertising). Further, such direct and indirect effects depend on TPR size. The findings indicate significant direct effects: the platform owner's price promotion has a positive spillover effect on TPR sales, and more so for larger TPRs; the platform owner's in-platform advertising has a negative crowding-out effect on TPR sales and more so for smaller TPRs. The paper also reports indirect effects: TPRs tend to use price promotion to respond to both strategies of the platform owner, and TPR's price promotion also has positive consequences for TPR sales, and the effects are stronger for larger TPRs. Given the unique coopetition relationship between the platform owner and TPRs in the hybrid retailing context, this research provides firsthand empirical evidence that the platform owner's strategies can significantly affect TPR performance through direct and indirect effects, shows larger TPRs benefit the most from the platform owner's price promotion, while smaller TPRs suffer the most from the platform owner's in-platform advertising, and TPRs favor price promotion regardless what strategies the platform owner uses. These results indicate that both platform owners and policymakers should consider the differential implications for TPRs of different sizes in terms of the social welfare of the platform ecosystem.

Longitudinal gambling risk transitions: evidence from a nationally representative Australian sample

Aims Problem gambling has downstream consequences on individuals, families and the community. While a strong research base now exists on predictors and outcomes of problem gambling risk and severity, few studies have examined the transitions in gambling risk status over time, and factors associated with these transitions. Methods The current study addresses this knowledge gap by examining gambling transitions using two waves of longitudinal, population-representative Australian data (N = 12,364) collected in 2015 and 2018. Focal to our approach is the assessment of predictors of gambling risk status including demographics, and key psychosocial factors that might attenuate the risk of transition into more severe gambling. Results The results show significant stability in gambling risk over time, particularly among individuals who reported no or low gambling risk. Gambling risk transitions were more likely to occur toward less severe than more severe levels of gambling. Furthermore, gambling problems tended to persist in more severe levels of gambling risk. Financial hardship, younger age, male gender, experience of hardship, lower levels of educational attainment, chronic health conditions, risky levels of alcohol consumption, living in low SES areas, and low sense of mastery were associated with transitions from low to more severe gambling over the three-year period. There were no significant predictors of transitions to lower levels of gambling risk in the current data. Conclusions Our findings can help inform public health interventions by better targeting individuals at elevated risk for more severe gambling over time, and we outline a method for analyzing transitions in longitudinal datasets that can be applied in future studies in addiction.

Examining neural responses to anticipating or receiving monetary rewards and the development of binge eating in youth. A registered report using data from the Adolescent Brain Cognitive Development (ABCD) study

Binge eating is characterized as eating a large amount of food and feeling a loss of control while eating. However, the neurobiological mechanisms associated with the onset and maintenance of binge eating are largely unknown. Recent neuroimaging work has suggested that increased responsivity within reward regions of the brain to the anticipation or receipt of rewards is related to binge eating; however, limited longitudinal data has precluded understanding of the role of reward responsivity in the development of binge eating. The current study used data from the Adolescent Brain and Cognitive Development® (ABCD) longitudinal study dataset to assess whether heightened neural responses to different phases of reward processing (reward anticipation and receipt) (1) differentiated individuals with binge eating from matched controls, and (2) predicted the onset of binge eating in an "at risk" sample. Consistent with hypotheses, heightened neural responsivity in the right caudate and bilateral VS during reward anticipation differentiated youth with and without binge eating. Moreover, greater VS response to reward anticipation predicted binge eating two years later. Neural responses to reward receipt also were consistent with hypotheses, such that heightened VS and OFC responses differentiated youth with and without binge eating and predicted the presence of binge eating two years later. Findings from the current study suggest that hypersensitivity to rewards may contribute to the development of binge eating during early adolescence.

Debate: CAMHS will be at the forefront of the next generation of psychosis risk models, but further integration with early intervention psychosis services is needed to realise this potential: Re Debate: Prevention of psychosis in adolescents - does CAMHS have a role?

The detection of psychosis and its prodrome have unique considerations in a child and adolescent population. Young people attending CAMHS are already a high-risk group, which confers significant limitations in applying the current clinical high-risk (CHR) model. This has catalysed calls for a transdiagnostic approach to psychosis risk prediction, but without a clear pathway forward. We contribute to the debate opened by Salazar de Pablo and Arango (2023, Child and Adolescent Mental Health) on the role of CAMHS in this initiative. CAMHS have a key role in developing comprehensive longitudinal datasets to inform risk models. Closer integration with early intervention in psychosis (EIP) services will be needed to realise this potential. This integration is also required to reliably detect prodromes and emerging psychosis in young people. Where there is robust evidence to support prevention initiatives, we should proceed with their implementation, even in the absence of enhanced risk models.

Seeking the roots of entrepreneurship: Childhood and adolescence extraversion predict entrepreneurial intention in adults

AbstractEntrepreneurial personality research has mostly used cross‐sectional data, giving rise to the question of whether the results can be used to predict entrepreneurial outcomes. Using a unique longitudinal dataset, we investigate whether and from what age personality in childhood and adolescence predicts entrepreneurial intention in young adulthood. We cover a time span of 17 years by assessing personality at four different developmental phases (6–9, 9–12, 12–15, and 14–17 years of age) and one measurement of entrepreneurial intention in young adulthood (23–26 years of age). We employ a variable‐oriented approach by studying the Big Five traits—imagination, conscientiousness, extraversion, benevolence, and emotional stability—individually and a person‐oriented approach by studying the “entrepreneurial personality profile,” a configuration of these traits. Extraversion as measured at 9–12 years already predicts entrepreneurial intention in young adulthood. This relationship becomes stronger as the age at which extraversion is measured increases. Unlike earlier literature, we do not find clear prospective relationships regarding the other traits. Our study presents insights into the antecedents of entrepreneurial intention and highlights the importance of extraversion in childhood and adolescence. The difference between our results and those of previous cross‐sectional research highlights the importance of using longitudinal data.

Parenting by mothers from marginalized communities and the role of socioeconomic disadvantage: insights from marginalized Roma communities in Slovakia.

Roma living in marginalized communities often face poor living conditions and material deprivation, which may negatively impact parenting. Our aim is to compare the parenting behavior (support, harsh discipline, and stimulation) of mothers from marginalized Roma communities and the majority population in Slovakia. We also examine the role of socioeconomic disadvantage and related worries in the differences in parenting behavior between these groups. We obtained cross-sectional data from mothers of children aged 14-18 months using the first wave of the longitudinal RomaREACH study dataset. Two groups were included in the sample: 93 mothers from MRCs and 102 mothers from the majority. We performed multiple regression and mediation analyses to assess whether the educational level of mothers, the degree of poverty, and poverty-related feelings of stress and worries explain parenting behavior differences between the groups of mothers. We found significant differences in parenting, especially in harsh disciplining and stimulation. These two domains were significantly associated with maternal education, degree of poverty, and poverty-related stress and worries. The degree of poverty partially mediated stimulation differences between the two groups of mothers. Parenting in MRCs seems harsher and less stimulative than parenting in the Slovak majority. These differences are associated with the socioeconomic disadvantage of mothers. The degree of poverty partially explains why parenting in MRCs is less stimulative. These results may inform intervention efforts aimed at disadvantaged families.

Multilevel Synergy of Information Technology for Operational Integration: Competition Networks and Operating Performance

Firms’ multilevel access to information plays a significant role in improving operating performance. Increasingly, firms are enhancing their operational integration through information technology (IT) as they grapple with intense competition. Competition networks are an essential but often overlooked source of information that, if leveraged correctly, can provide firms with significant competitive and operational advantages. In this study, we develop a multilevel research model of operating performance (firm level) that simultaneously considers the effects of IT for operational integration (ITOI, firm level) and competitive brokerage (competition network level). We explicate how ITOI and competitive brokerage afford firms’ synergy through complementarities and relatedness of competitive actions, information, and resources to improve their operating performance. We assess the model using a 7-year longitudinal secondary data set that includes firms from multiple industries and find support for our thesis that ITOI and competitive brokerage have a synergistic effect on operating performance. Our exploratory analysis uncovers innovation efficiency as a theoretical mechanism underlying the relationships between ITOI, competitive brokerage, and operating performance. Further exploratory analyses with disaggregated measures of ITOI highlight that synergies arise from IT-enabled coordination and integration across the supply chain and within functional areas of an organization. These findings are robust to concerns of endogeneity and alternative model specifications. We make three critical contributions to our collective understanding of the relationship between IT and operating performance—synergy as a means through which operating performance is realized, a multilevel model of relationships, and a nuanced understanding of the relationships between ITOI, competitive brokerage, and operating performance. Overall, we provide a summative view of the multilevel effects of IT on operating performance.

Changes in efficiency and physical size of container ports: An integration of genetic matching and stochastic data envelopment analysis

Benchmarking container ports of different physical sizes and figuring out the relationship between size and efficiency is complex due to the heterogeneous environment. In dealing with heterogeneity, the selection bias is often overlooked. Therefore, this study proposes an integrated multivariate genetic matching and stochastic DEA algorithm to evaluate the efficiency of container ports. It paves the way for container ports that differ in a selected feature, such as size, to be benchmarked using well-balanced clusters. Thus, the port managers can identify the most similar-featured peers in benchmarking with DEA or alternative models to acquire robust estimates without dependence on a longitudinal data set. The results of the model applied to international container ports imply the increase in size impacts efficiency negatively, while connectivity does positively, which contradicts the commonly held perception of stakeholders that the larger the container ports, the more efficient. That is, well-managed small container ports can also be as efficient. Therefore, it is concluded that the results of integrating genetic matching into the performance measurement provide beneficial inferences. In future research, clustering the observed container ports according to a specific feature and balancing clusters with multivariate genetic matching can provide valuable insights into the industry.

Horse populations are severely underestimated in a region at risk of Hendra virus spillover.

To identify the size and distribution of the horse population in the Northern Rivers Region of NSW, including changes from 2007 to 2021, to better understand populations at risk of Hendra virus transmission. Census data from the 2007 Equine Influenza (EI) outbreak were compared with data collected annually by New South Wales Local Land Services (LLS) (2011-2021), and with field observations via road line transects (2021). The horse populations reported to LLS in 2011 (3000 horses; 0.77 horses/km2) was 145% larger than that reported during the EI outbreak in 2007 (1225 horses; 0.32 horses/km2). This was inconsistent with the 6% increase in horses recorded from 2011 to 2020 within the longitudinal LLS dataset. Linear modelling suggested the true horse population of this region in 2007 was at least double that reported at the time. Distance sampling in 2021 estimated the region's population at 10,185 horses (3.89 per km2; 95% CI = 4854-21,372). Field sampling and modelling identified higher horse densities in rural cropland, with the percentage of conservation land, modified grazing, and rural residential land identified as the best predictors of horse densities. Data from the 2007 EI outbreak no longer correlates to the current horse population in size or distribution and was likely not a true representation at the time. Current LLS data also likely underestimates horse populations. Ongoing efforts to further quantify and map horse populations in Australia are important for estimating and managing the risk of equine zoonoses.

Longitudinal Analysis of Latvian Child Growth: Anthropometric Parameters Dynamics from Birth to Adolescence.

This study provides a comprehensive analysis of the physical development patterns from birth to adolescence, utilizing a longitudinal dataset of 70 children monitored from birth until 17 years of age. The research focuses on the variability of growth trajectories, emphasizing the role of genetic and environmental factors in influencing these patterns. Key findings indicate that most children undergo one or two periods of accelerated growth, with significant variability in the timing and magnitude of these growth spurts. The study also highlights the adaptive nature of growth changes over generations, influenced by ecological, nutritional, and socio-economic conditions. The longitudinal approach reveals critical insights into the timing of peak growth velocities, demonstrating that girls reach their growth peak approximately one year earlier than boys. The analysis of intergenerational growth patterns suggests a significant increase in average height over the century, attributed to genetic diversity and changes in lifestyle and nutrition. This study's findings emphasize the importance of updating physical development standards regularly to reflect the changing genetic and environmental landscape. The variability in growth patterns and their correlation with health outcomes in later life highlights the need for targeted public health strategies that address the underlying socio-economic and environmental determinants of health. This research contributes to the understanding of physical development trajectories and provides a foundation for future studies aimed at optimizing health outcomes from early childhood through adolescence. The primary objective of this article is to meticulously analyze the dynamics of height growth and accurately identify the periods of accelerated bodily development within the context of longitudinal research.

A milestone in the pursuit of gender equality: Predicting first women presidents in U.S. higher education institutions, 1980–2018

AbstractWhile women’s higher education enrollments and graduation rates have outpaced those of men in the United States and most countries around the world, women are less frequently included in academic leadership roles, including the higher education presidency. This paper asks what predicts whether and when a higher education institution has its first woman president, conceptualizing this event as a milestone of gender equality. We use a national probability sample of 234 four‐year U.S. universities and colleges, constructing a novel longitudinal dataset from 1980 to 2018. Employing event history analysis, we examine the potential mechanisms associated with when an institution has its first woman president over time. Our findings suggest that the demographic diversity of faculty and students, gender‐ and diversity‐supportive structures, and the broader environment in which institutions are embedded predict the likelihood that a woman will advance to the level of the presidency. In particular, the presence of gender studies programs and a higher proportion of women in state legislatures increase the likelihood that an institution will have its first woman president. At a time of growing challenges facing U.S. higher education, coupled with greater opportunities from having more diverse students and faculty, universities and colleges increasingly recognize the benefit of women leaders.

Using Natural Language Processing to Identify Home Health Care Patients at Risk for Diagnosis of Alzheimer's Disease and Related Dementias.

This study aimed to: (1) validate a natural language processing (NLP) system developed for the home health care setting to identify signs and symptoms of Alzheimer's disease and related dementias (ADRD) documented in clinicians' free-text notes; (2) determine whether signs and symptoms detected via NLP help to identify patients at risk of a new ADRD diagnosis within four years after admission. This study applied NLP to a longitudinal dataset including medical record and Medicare claims data for 56,652 home health care patients and Cox proportional hazard models to the subset of 24,874 patients admitted without an ADRD diagnosis. Selected ADRD signs and symptoms were associated with increased risk of a new ADRD diagnosis during follow-up, including: motor issues; hoarding/cluttering; uncooperative behavior; delusions or hallucinations; mention of ADRD disease names; and caregiver stress. NLP can help to identify patients in need of ADRD-related evaluation and support services.

Longitudinal Patterns of Online Activity and Social Feedback Are Associated with Current and Perceived Changes in Quality of Life in Adult Facebook Users

The present study explored how sharing verbal status updates on Facebook and receiving Likes, as a form of positive social feedback, correlate with current and perceived changes in Quality of Life (QoL). Utilizing the Facebook Graph API, we collected a longitudinal dataset comprising status updates and Likes received by 1577 adult Facebook users over a 12-month period. Two monthly indicators were calculated: the percentage of verbal status updates and the average number of Likes per post. Participants were administered a survey to assess current and perceived changes in QoL. Confirmatory Factor Analysis (CFA) and the Auto-Regressive Latent Trajectory Model with Structured Residuals (ALT-SRs) were used to model longitudinal patterns emerging from the objective recordings of Facebook activity and explore their correlation with QoL measures. Findings indicated a positive correlation between the percentage of verbal status updated on Facebook and current QoL. Online positive social feedback, measured through received Likes, was associated with both current QoL and perceived improvements in QoL. Of note, perceived improvements in QoL correlated with an increase in received Likes over time. Results highlight the relevance of collecting and modeling longitudinal Facebook data for the investigation of the association between activity on social media and individual well-being.

A four-cohort study testing the relative impact of take-home and in-class examination on students’ academic performance and wellbeing

The current study examines the relative impact of take home (open book) examinations (THE) and in class (closed book) examinations (ICE) on student academic performance and wellbeing outcomes. Specifically, this study contributes to our understanding about students’ long-term knowledge retention, measured four to six months after completing the course. A unique longitudinal dataset consisting of four cohorts of a social science bachelor and master course that implemented either a THE or ICE in successive years was used. Survey data included wellbeing and academic performance measures as well as a 10-item knowledge retention quiz. Within the master course, the ICE cohort had higher examination grades and higher knowledge retention scores than the cohorts that completed a THE. In the bachelor course, there were no differences in knowledge retention across cohorts. Examination score was associated with higher knowledge retention scores across both courses. One bachelor cohort reported lower wellbeing compared to others (cohort 2021–2022); however, we found no further differences in academic or wellbeing outcomes based on examination form. The findings suggest a slight advantage of ICEs over THEs with regard to academic performance.

Non-steroidal anti-inflammatory drugs, renin-angiotensin system blockade or diuretics and risk of acute kidney injury: A case-crossover study

BackgroundAging-related physiological changes, such as decline in renal function, not only exacerbates pre-existing comorbidities but also escalate the susceptibility to adverse events. Previous studies have shown that non-steroidal anti-inflammatory drugs (NSAIDs) are associated with an increased risk of acute kidney injury (AKI), and the concomitant use of renin-angiotensin system blockade or diuretics may further potentiate the risk. However, studies evaluating the risk of AKI associated with NSAIDs (including routes, concomitant use of different NSAIDs, categories (traditional NSAIDs or COX-2 inhibitors), and cumulative doses of NSAIDs) are limited, particularly the risk of AKI associated with the dual or triple combination of NSAIDs with renin-angiotensin system blockade (RAS blockades) and/or diuretics. MethodsA case-crossover study utilized two sets of longitudinal data from Taiwan's National Health Insurance Research Database (NHIRD). Newly admitted patients with a primary AKI diagnosis were included, with the index date defined as the first admission date. The 1–7 days and 181–187 days prior to the index date served as the case and control periods. Exposure to NSAIDs and co-exposures of RAS blockade and/or diuretics were assessed in both periods. Multivariable conditional logistic regression models, adjusting for potential confounders, estimated adjusted odds ratios (aORs) and 95 % confidence intervals (CIs) for AKI associated with NSAIDs, dual, or triple combinations. Sensitivity analyses explored result robustness by varying case and control period lengths. ResultsThe study included 1,284 newly diagnosed AKI patients. NSAIDs showed a 3.55-fold increased risk of AKI (aOR: 3.55; 95 % CI 2.70–4.65), with similar risks for traditional NSAIDs and COX-2 inhibitors. Use of multiple NSAIDs, parenteral dosage forms, and higher cumulative doses increased AKI risk. Dual combination with either RAS blockade or diuretics resulted in a 2.90-fold (aOR: 2.90; 95 %CI 1.47–5.70) and 12.68-fold (aOR: 12.68; 95 %CI 6.15–26.12) risk, respectively. The highest risk occurred with triple combination (aOR: 29.22; 95 %CI 12.82–66.64). ConclusionsNSAIDs, including both non-selective NSAIDs and COX2 inhibitors, elevate the risk of AKI. Increased AKI risk is linked to using multiple NSAIDs, the parenteral dosage form, and higher cumulative doses. Dual combination of RAS blockade with NSAIDs or diuretics with NSAIDs, as well as triple therapy, heightens the risk, with the latter associated with the highest risk of AKI.

Benefits of Pair-Bond Duration on Reproduction in a Lifelong Monogamous Cooperative Passerine.

AbstractLong-term social and genetic monogamy is rare in animals except birds, but even in birds it is infrequent and poorly understood. We investigated possible advantages of monogamy in a colonial, facultative cooperatively breeding bird from an arid, unpredictable environment, the sociable weaver (Philetairus socius). We documented divorce and extrapair paternity of 703 pairs over 10 years and separated effects of pair duration from breeding experience by analyzing longitudinal and cross-sectional datasets. Parts of the colonies were protected from nest predation, thereby limiting its stochastic and thus confounding effect on fitness measures. We found that 6.4% of sociable weaver pairs divorced and 2.2% of young were extrapair. Longer pair-bonds were associated with more clutches and fledglings per season and with reproducing earlier and later in the season, when snake predation is lower, but not with increased egg or fledgling mass or with nestling survival. Finally, the number of helpers at the nest increased with pair-bond duration. Results were similar for protected and unprotected nests. We suggest that long-term monogamy is associated with a better capacity for exploiting a temporally unpredictable environment and helps to form larger groups. These results can contribute to our understanding of why long-term monogamy is frequently associated with unpredictable environments and cooperation.

Conditional Generative Adversarial Network for Early Classification of Longitudinal Datasets Using an Imputation Approach

Early classification of longitudinal data remains an active area of research today. The complexity of these datasets and the high rates of missing data caused by irregular sampling present data-level challenges for the Early Longitudinal Data Classification (ELDC) problem. Coupled with the algorithmic challenge of optimising the opposing objectives of early classification (i.e., earliness and accuracy), ELDC becomes a non-trivial task. Inspired by the generative power and utility of the Generative Adversarial Network (GAN), we propose a novel context-conditional, longitudinal early classifier GAN (LEC-GAN). This model utilises informative missingness, static features and earlier observations to improve the ELDC objective. It achieves this by incorporating ELDC as an auxiliary task within an imputation optimization process. Our experiments on several datasets demonstrate that LEC-GAN outperforms all relevant baselines in terms of F1 scores while increasing the earliness of prediction.

Machine learning with real-world HR data: mitigating the trade-off between predictive performance and transparency

Machine Learning (ML) algorithms offer a powerful tool for capturing multifaceted relationships through inductive research to gain insights and support decision-making in practice. This study contributes to understanding the dilemma whereby the more complex ML becomes, the more its value proposition can be compromised by its opacity. Using a longitudinal dataset on voluntary employee turnover from a German federal agency, we provide evidence for the underlying trade-off between predictive performance and transparency for ML, which has not been found in similar Human Resource Management (HRM) studies using artificially simulated datasets. We then propose measures to mitigate this trade-off by demonstrating the use of post-hoc explanatory methods to extract local (employee-specific) and global (organisation-wide) predictor effects. After that, we discuss their limitations, providing a nuanced perspective on the circumstances under which the use of post-hoc explanatory methods is justified. Namely, when a ‘transparency-by-design’ approach with traditional linear regression is not sufficient to solve HRM prediction tasks, the translation of complex ML models into human-understandable visualisations is required. As theoretical implications, this paper suggests that we can only fully understand the multi-layered HR phenomena explained to us by real-world data if we incorporate ML-based inductive methods together with traditional deductive methods.

Abstract 6223: Personalized MRI-informed forecasting of prostate cancer progression during active surveillance

Abstract Low and intermediate-risk prostate cancer (PCa) patients are eligible for active surveillance (AS), whereby treatment is delayed until progression to higher-risk disease is detected via longitudinal multiparametric magnetic resonance imaging (mpMRI) scans, biopsies, and prostate specific antigen (PSA) tests. However, AS relies on a population-based observational paradigm in which standard testing frequencies may not match the patient’s tumor dynamics, resulting in a delayed diagnosis of progressive disease. To address these issues, we propose to advance AS towards a patient-specific predictive paradigm by leveraging computational tumor forecasts obtained with a biomechanistic model informed by mpMRI and clinical data collected during standard AS. Here, we present a retrospective study with n = 16 PCa cases. All patients had three mpMRI scans obtained over 2.6 to 5.6 years during AS. Our biomechanistic model describes PCa growth in terms of the dynamics of tumor cell density as a combination of tumor cell mobility and net proliferation. The model is defined on the patient’s prostate geometry, which is segmented on T2-weighted MRI data. Tumor cell density estimates are derived from apparent diffusion coefficient (ADC) maps obtained from diffusion-weighted MRI data. To facilitate modeling, the longitudinal imaging datasets are non-rigidly co-registered for each patient. We initialize the model with the tumor cell density map obtained from the ADC map of the first scan. The model is parameterized by minimizing the model-data mismatch in tumor cell density at the second scan date, and then we perform a tumor forecast up to the third scan date. Finally, we build a logistic classifier from a panel of model-based biomarkers calculated from the personalized model forecasts (e.g., tumor volume, total proliferation activity) at the times of histopathological assessment (i.e., biopsy, surgery) to classify tumors as low risk (Gleason score 3+3, n = 16) or intermediate-high risk (Gleason score ≥ 3+4, n = 15). We obtained a concordance correlation coefficient (CCC) for tumor volume of 0.89 at both model calibration and forecasting horizon. The spatial fit of tumor cell density yielded a median Dice score and CCC of 0.80 and 0.59 at the second mpMRI date, and of 0.76 and 0.60 at the third mpMRI date, respectively. The logistic classifier of PCa risk yielded an area under the ROC curve of 0.90 and operates at optimal sensitivity, specificity, and accuracy of 86.7%, 93.8%, and 90.3%, respectively. The time trajectories of PCa risk obtained from the temporal predictions of the model-based biomarkers enabled identification of PCa progression by more than 1 year. Thus, while further development and validation over larger cohorts are required, these results suggest that our forecasting technology is a promising tool to predict PCa progression in AS and, hence, optimally guide monitoring and treatment decisions during AS. Citation Format: Guillermo Lorenzo Gomez, Chengyue Wu, Joshua P. Yung, John F. Ward, Hector Gomez, Alessandro Reali, Thomas E. Yankeelov, Aradhana M. Venkatesan, Thomas J. Hughes. Personalized MRI-informed forecasting of prostate cancer progression during active surveillance [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6223.

Improved clinical outcome prediction in depression using neurodynamics in an emotional face-matching functional MRI task.

Approximately one in six people will experience an episode of major depressive disorder (MDD) in their lifetime. Effective treatment is hindered by subjective clinical decision-making and a lack of objective prognostic biomarkers. Functional MRI (fMRI) could provide such an objective measure but the majority of MDD studies has focused on static approaches, disregarding the rapidly changing nature of the brain. In this study, we aim to predict depression severity changes at 3 and 6 months using dynamic fMRI features. For our research, we acquired a longitudinal dataset of 32 MDD patients with fMRI scans acquired at baseline and clinical follow-ups 3 and 6 months later. Several measures were derived from an emotion face-matching fMRI dataset: activity in brain regions, static and dynamic functional connectivity between functional brain networks (FBNs) and two measures from a wavelet coherence analysis approach. All fMRI features were evaluated independently, with and without demographic and clinical parameters. Patients were divided into two classes based on changes in depression severity at both follow-ups. The number of coherence clusters (nCC) between FBNs, reflecting the total number of interactions (either synchronous, anti-synchronous or causal), resulted in the highest predictive performance. The nCC-based classifier achieved 87.5% and 77.4% accuracy for the 3- and 6-months change in severity, respectively. Furthermore, regression analyses supported the potential of nCC for predicting depression severity on a continuous scale. The posterior default mode network (DMN), dorsal attention network (DAN) and two visual networks were the most important networks in the optimal nCC models. Reduced nCC was associated with a poorer depression course, suggesting deficits in sustained attention to and coping with emotion-related faces. An ensemble of classifiers with demographic, clinical and lead coherence features, a measure of dynamic causality, resulted in a 3-months clinical outcome prediction accuracy of 81.2%. The dynamic wavelet features demonstrated high accuracy in predicting individual depression severity change. Features describing brain dynamics could enhance understanding of depression and support clinical decision-making. Further studies are required to evaluate their robustness and replicability in larger cohorts.