Digital phenotyping in young breast cancer patients treated with neoadjuvant chemotherapy (the NeoFit Trial): protocol for a national, multicenter single-arm trial

Delrieu, Lidia; Hamy, Anne-Sophie; Kassara, Amyn; Asselain, Bernard; Antero, Juliana; Villèle, Paul De; Dumas, Élise; Forstmann, Nicolas; Guérin, J.; Hotton, Judicaël; Jouannaud, Christelle; Milder, Maud; Leopold, Armand; Sedeaud, Adrien; Soibinet, Pauline; Toussaint, Jean‐François; Vercamer, Vincent; Laas, Enora; Reyal, Fabien

doi:10.1186/s12885-022-09608-y

Cited by 3 publications

(2 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…30 Based on similar work on digital phenotyping, this sample size and amount of data are sufficient to build reliable models. While some publications used more patients (eg, 16,552 unique patients to phenotype sepsis 23 ; 134,252 for arthroplasties 27 ), other authors published relevant results based on a smaller samples of patients (eg, 608 patients for coronavirus disease-2019 [COVID-19] 19 ; 300 patients for breast cancer, 31 105 patients for spine disease 32 ).…”

Section: Databasementioning

confidence: 99%

Novel Preoperative Risk Stratification Using Digital Phenotyping Applying a Scalable Machine-Learning Approach

Laferrière-Langlois,

Imrie,

Geraldo

et al. 2023

Anesthesia &Amp; Analgesia

View full text Add to dashboard Cite

BACKGROUND: Classification of perioperative risk is important for patient care, resource allocation, and guiding shared decision-making. Using discriminative features from the electronic health record (EHR), machine-learning algorithms can create digital phenotypes among heterogenous populations, representing distinct patient subpopulations grouped by shared characteristics, from which we can personalize care, anticipate clinical care trajectories, and explore therapies. We hypothesized that digital phenotypes in preoperative settings are associated with postoperative adverse events including in-hospital and 30-day mortality, 30-day surgical redo, intensive care unit (ICU) admission, and hospital length of stay (LOS). METHODS: We identified all laminectomies, colectomies, and thoracic surgeries performed over a 9-year period from a large hospital system. Seventy-seven readily extractable preoperative features were first selected from clinical consensus, including demographics, medical history, and lab results. Three surgery-specific datasets were built and split into derivation and validation cohorts using chronological occurrence. Consensus k-means clustering was performed independently on each derivation cohort, from which phenotypes’ characteristics were explored. Cluster assignments were used to train a random forest model to assign patient phenotypes in validation cohorts. We reconducted descriptive analyses on validation cohorts to confirm the similarity of patient characteristics with derivation cohorts, and quantified the association of each phenotype with postoperative adverse events by using the area under receiver operating characteristic curve (AUROC). We compared our approach to American Society of Anesthesiologists (ASA) alone and investigated a combination of our phenotypes with the ASA score. RESULTS: A total of 7251 patients met inclusion criteria, of which 2770 were held out in a validation dataset based on chronological occurrence. Using segmentation metrics and clinical consensus, 3 distinct phenotypes were created for each surgery. The main features used for segmentation included urgency of the procedure, preoperative LOS, age, and comorbidities. The most relevant characteristics varied for each of the 3 surgeries. Low-risk phenotype alpha was the most common (2039 of 2770, 74%), while high-risk phenotype gamma was the rarest (302 of 2770, 11%). Adverse outcomes progressively increased from phenotypes alpha to gamma, including 30-day mortality (0.3%, 2.1%, and 6.0%, respectively), in-hospital mortality (0.2%, 2.3%, and 7.3%), and prolonged hospital LOS (3.4%, 22.1%, and 25.8%). When combined with the ASA score, digital phenotypes achieved higher AUROC than the ASA score alone (hospital mortality: 0.91 vs 0.84; prolonged hospitalization: 0.80 vs 0.71). CONCLUSIONS: For 3 frequently performed surgeries, we identified 3 digital phenotypes. The typical profiles of each phenotype were described and could be used to anticipate adverse postoperative events.

show abstract

Section: Databasementioning

confidence: 99%

Novel Preoperative Risk Stratification Using Digital Phenotyping Applying a Scalable Machine-Learning Approach

Laferrière-Langlois,

Imrie,

Geraldo

et al. 2023

Anesthesia &Amp; Analgesia

View full text Add to dashboard Cite

show abstract

“…In essence, these advancements in technology pave the way for a more nuanced understanding of individuals' PA patterns and contribute to more accurate and insightful clinical assessments (23,24,25,26). This evolution has given rise to a concept known as digital phenotyping (DP), which refers to the utilization of data sourced from personal digital devices to swiftly quantify the individual human phenotype (27). DP achieves real-time, continuous quanti cation in an unsupervised way of an individual within their natural environment through the automated aggregation of data.…”

Section: Introductionmentioning

confidence: 99%

Unveiling the digital phenotype: A protocol for a prospective study on physical activity behavior in community-dwelling older adults

Daniels,

Vonck,

Robijns

et al. 2024

Preprint

View full text Add to dashboard Cite

Background. Physical activity (PA) is a recognized boon for older adults, enhancing their overall well-being and mitigating health risks. Nevertheless, to encourage active lifestyles in this demographic, it is vital to understand the factors influencing PA. Conventional approaches predominantly rely on supervised cross-sectional evaluations, presuming both the stability of PA determinants over time and their isolated components. However, the complex nature of real-life dynamics often involves temporal variability in individual-level determinants. Digital phenotyping (DP), employing data recruited from personal digital devices, enables the continuous, unsupervised and real-time quantification of an individual's behavior within their natural context. This approach offers more ecological and dynamic assessments, revolutionizing our understanding of the intricacies underlying individual PA patterns in their environmental context. Objective. This paper aims to design a robust research protocol for the DP of PA behavior among healthy community-dwelling older adults aged 65 and above by employing a novel measurement approach. Methods. Observational data will be collected over a two-week period to assess various functions combining both cross-sectional and longitudinal data collection methods. Patterns of PA behavior and factors affecting PA outcomes will be detected in order to identify digital phenotypes related to PA. The measurements are based on the Behavior Change Wheel and include self-reporting and clinical assessments for cross-sectional data collection and ecological momentary assessment as well as time series collection for longitudinal data. The statistical analysis involves machine learning which will handle data complexity. Unsupervised learning will be used to uncover patterns, and supervised learning to identify variables. The analysis will be conducted in RStudio (v3.6.3) with significance set at 0.05. Discussion. A novel approach to understanding older adults' PA behavior will be used in this study. Challenges include varying technology adoption, usability, and unproven validity of health tech. Ethical considerations, representativeness, participant engagement, and machine learning expertise are also key aspects of the study's success. This study offers promise in bridging traditional and dynamic assessment methods for older adults' PA behavior to promote active lifestyles. Trial registration: Clinical Trials.gov: NCT06094374

show abstract

A Behavioral Perspective for Improving Exercise Adherence

André,

Grousset,

Audiffren

2024

Sports Med - Open

View full text Add to dashboard Cite

Adherence to exercise is a crucial aspect of behavioral changes related to exercise. However, current models fail to predict behavioral change, and exercise programs struggle to foster adherence. In this Current Opinion article, we defined adherence as a process linking behavior and attitude. More specifically, exercise adherence is the process by which people’s behaviors conform to their attitudes and vice versa. Behavioral change theories should be able to predict changes, transformations, and modifications of behaviors; however, this is not currently the case. Prediction models of behavior are mainly focused on how to predict behavioral changes by affecting attitudes; however, these models have not considered the target behavior as a facilitator of adherence. Herein, a behavioral perspective is considered in three directions: first, exercise is a therapeutic modality that has the potential to improve motivation; second, chronic exercise helps sustain effort over time and increase executive functions and willpower; and third, exercise is an active behavior that has the potential to facilitate attitude change.

show abstract

Digital phenotyping in young breast cancer patients treated with neoadjuvant chemotherapy (the NeoFit Trial): protocol for a national, multicenter single-arm trial

Cited by 3 publications

References 53 publications

Novel Preoperative Risk Stratification Using Digital Phenotyping Applying a Scalable Machine-Learning Approach

Novel Preoperative Risk Stratification Using Digital Phenotyping Applying a Scalable Machine-Learning Approach

Unveiling the digital phenotype: A protocol for a prospective study on physical activity behavior in community-dwelling older adults

A Behavioral Perspective for Improving Exercise Adherence

Contact Info

Product

Resources

About