Jena Daniels
Research Subjects
Research Interests
Jena Daniels has a total of 37 co-authors. Here are the top co-authors by number of shared publications:
- Dennis Paul Wall
Department of Pediatrics, Division of Clinical Informatics, Stanford University School of Medicine, Palo Alto, CA 94305, USA
7 shared publications - Jessey Schwartz
Departments of Pediatrics, Biomedical Data Science, Psychiatry and Behavioral Sciences, Stanford University, Stanford, CA, United States
6 shared publications - Peter Washington
Division of Clinical Informatics and Digital Transformation (DoC-IT), Department of Medicine, University of California, San Francisco, USA
4 shared publications - Michael Du
Division of Systems Medicine, Department of Pediatrics, Stanford University, Stanford, CA, United States
3 shared publications - Allison W. Kurian
Department of Medicine, Stanford University School of Medicine , Stanford, CA
2 shared publications - Anish Nag
University of California, Berkeley, CA, United States
2 shared publications
Latest Publications
- Open Access
- 13 August 2021
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 5  (8) , e22608
Development of a Mobile Health App (TOGETHERCare) to Reduce Cancer Care Partner Burden: Product Design Study
Background: Approximately 6.1 million adults in the United States serve as care partners for cancer survivors. Studies have demonstrated that engaging cancer survivors and their care partners through technology-enabled structured symptom collection has several benefits. Given the high utilization of mobile technologies, even among underserved populations and in low resource areas, mobile apps may provide a meaningful access point for all stakeholders for symptom management. Objective: We aimed to develop a mobile app incorporating user preferences to enable cancer survivors’ care partners to monitor the survivors’ health and to provide care partner resources. Methods: An iterative information gathering process was conducted that included (1) discussions with 138 stakeholders to identify challenges and gaps in survivor home care; (2) semistructured interviews with clinicians (n=3), cancer survivors (n=3), and care partners (n=3) to identify specific needs; and (3) a 28-day feasibility field test with seven care partners. Results: Health professionals noted the importance of identifying early symptoms of adverse events. Survivors requested modules on medication, diet, self-care, reminders, and a version in Spanish. Care partners preferred to focus primarily on the patient’s health and not their own. The app was developed incorporating quality-of-life surveys and symptom reporting, as well as resources on home survivor care. Early user testing demonstrated ease of use and app feasibility. Conclusions: TOGETHERCare, a novel mobile app, was developed with user input to track the care partner’s health and report on survivor symptoms during home care. The following two clinical benefits emerged: (1) reduced anxiety among care partners who use the app and (2) the potential for identifying survivor symptoms noted by the care partner, which might prevent adverse events. Trial Registration: ClinicalTrials.gov NCT04018677; https://clinicaltrials.gov/ct2/show/NCT04018677
- 13 August 2021
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 5  (8) , e22608
Open AccessDevelopment of a Mobile Health App (TOGETHERCare) to Reduce Cancer Care Partner Burden: Product Design Study
Background: Approximately 6.1 million adults in the United States serve as care partners for cancer survivors. Studies have demonstrated that engaging cancer survivors and their care partners through technology-enabled structured symptom collection has several benefits. Given the high utilization of mobile technologies, even among underserved populations and in low resource areas, mobile apps may provide a meaningful access point for all stakeholders for symptom management. Objective: We aimed to develop a mobile app incorporating user preferences to enable cancer survivors’ care partners to monitor the survivors’ health and to provide care partner resources. Methods: An iterative information gathering process was conducted that included (1) discussions with 138 stakeholders to identify challenges and gaps in survivor home care; (2) semistructured interviews with clinicians (n=3), cancer survivors (n=3), and care partners (n=3) to identify specific needs; and (3) a 28-day feasibility field test with seven care partners. Results: Health professionals noted the importance of identifying early symptoms of adverse events. Survivors requested modules on medication, diet, self-care, reminders, and a version in Spanish. Care partners preferred to focus primarily on the patient’s health and not their own. The app was developed incorporating quality-of-life surveys and symptom reporting, as well as resources on home survivor care. Early user testing demonstrated ease of use and app feasibility. Conclusions: TOGETHERCare, a novel mobile app, was developed with user input to track the care partner’s health and report on survivor symptoms during home care. The following two clinical benefits emerged: (1) reduced anxiety among care partners who use the app and (2) the potential for identifying survivor symptoms noted by the care partner, which might prevent adverse events. Trial Registration: ClinicalTrials.gov NCT04018677; https://clinicaltrials.gov/ct2/show/NCT04018677
- Open Access
- 20 July 2020
- Published by JMIR Publications Inc. in JMIR Preprints
Development of a Mobile Health App (TOGETHERCare) to Reduce Cancer Care Partner Burden: Product Design Study (Preprint)
BACKGROUND Approximately 6.1 million adults in the United States serve as care partners for cancer survivors. Studies have demonstrated that engaging cancer survivors and their care partners through technology-enabled structured symptom collection has several benefits. Given the high utilization of mobile technologies, even among underserved populations and in low resource areas, mobile apps may provide a meaningful access point for all stakeholders for symptom management. OBJECTIVE We aimed to develop a mobile app incorporating user preferences to enable cancer survivors’ care partners to monitor the survivors’ health and to provide care partner resources. METHODS An iterative information gathering process was conducted that included (1) discussions with 138 stakeholders to identify challenges and gaps in survivor home care; (2) semistructured interviews with clinicians (n=3), cancer survivors (n=3), and care partners (n=3) to identify specific needs; and (3) a 28-day feasibility field test with seven care partners. RESULTS Health professionals noted the importance of identifying early symptoms of adverse events. Survivors requested modules on medication, diet, self-care, reminders, and a version in Spanish. Care partners preferred to focus primarily on the patient’s health and not their own. The app was developed incorporating quality-of-life surveys and symptom reporting, as well as resources on home survivor care. Early user testing demonstrated ease of use and app feasibility. CONCLUSIONS TOGETHERCare, a novel mobile app, was developed with user input to track the care partner’s health and report on survivor symptoms during home care. The following two clinical benefits emerged: (1) reduced anxiety among care partners who use the app and (2) the potential for identifying survivor symptoms noted by the care partner, which might prevent adverse events. CLINICALTRIAL ClinicalTrials.gov NCT04018677; https://clinicaltrials.gov/ct2/show/NCT04018677
- 20 July 2020
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessDevelopment of a Mobile Health App (TOGETHERCare) to Reduce Cancer Care Partner Burden: Product Design Study (Preprint)
BACKGROUND Approximately 6.1 million adults in the United States serve as care partners for cancer survivors. Studies have demonstrated that engaging cancer survivors and their care partners through technology-enabled structured symptom collection has several benefits. Given the high utilization of mobile technologies, even among underserved populations and in low resource areas, mobile apps may provide a meaningful access point for all stakeholders for symptom management. OBJECTIVE We aimed to develop a mobile app incorporating user preferences to enable cancer survivors’ care partners to monitor the survivors’ health and to provide care partner resources. METHODS An iterative information gathering process was conducted that included (1) discussions with 138 stakeholders to identify challenges and gaps in survivor home care; (2) semistructured interviews with clinicians (n=3), cancer survivors (n=3), and care partners (n=3) to identify specific needs; and (3) a 28-day feasibility field test with seven care partners. RESULTS Health professionals noted the importance of identifying early symptoms of adverse events. Survivors requested modules on medication, diet, self-care, reminders, and a version in Spanish. Care partners preferred to focus primarily on the patient’s health and not their own. The app was developed incorporating quality-of-life surveys and symptom reporting, as well as resources on home survivor care. Early user testing demonstrated ease of use and app feasibility. CONCLUSIONS TOGETHERCare, a novel mobile app, was developed with user input to track the care partner’s health and report on survivor symptoms during home care. The following two clinical benefits emerged: (1) reduced anxiety among care partners who use the app and (2) the potential for identifying survivor symptoms noted by the care partner, which might prevent adverse events. CLINICALTRIAL ClinicalTrials.gov NCT04018677; https://clinicaltrials.gov/ct2/show/NCT04018677
- Open Access
- 22 April 2020
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 22  (4) , e13810
Toward Continuous Social Phenotyping: Analyzing Gaze Patterns in an Emotion Recognition Task for Children With Autism Through Wearable Smart Glasses
Background: Several studies have shown that facial attention differs in children with autism. Measuring eye gaze and emotion recognition in children with autism is challenging, as standard clinical assessments must be delivered in clinical settings by a trained clinician. Wearable technologies may be able to bring eye gaze and emotion recognition into natural social interactions and settings. Objective: This study aimed to test: (1) the feasibility of tracking gaze using wearable smart glasses during a facial expression recognition task and (2) the ability of these gaze-tracking data, together with facial expression recognition responses, to distinguish children with autism from neurotypical controls (NCs). Methods: We compared the eye gaze and emotion recognition patterns of 16 children with autism spectrum disorder (ASD) and 17 children without ASD via wearable smart glasses fitted with a custom eye tracker. Children identified static facial expressions of images presented on a computer screen along with nonsocial distractors while wearing Google Glass and the eye tracker. Faces were presented in three trials, during one of which children received feedback in the form of the correct classification. We employed hybrid human-labeling and computer vision–enabled methods for pupil tracking and world–gaze translation calibration. We analyzed the impact of gaze and emotion recognition features in a prediction task aiming to distinguish children with ASD from NC participants. Results: Gaze and emotion recognition patterns enabled the training of a classifier that distinguished ASD and NC groups. However, it was unable to significantly outperform other classifiers that used only age and gender features, suggesting that further work is necessary to disentangle these effects. Conclusions: Although wearable smart glasses show promise in identifying subtle differences in gaze tracking and emotion recognition patterns in children with and without ASD, the present form factor and data do not allow for these differences to be reliably exploited by machine learning systems. Resolving these challenges will be an important step toward continuous tracking of the ASD phenotype.
- 22 April 2020
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 22  (4) , e13810
Open AccessToward Continuous Social Phenotyping: Analyzing Gaze Patterns in an Emotion Recognition Task for Children With Autism Through Wearable Smart Glasses
Background: Several studies have shown that facial attention differs in children with autism. Measuring eye gaze and emotion recognition in children with autism is challenging, as standard clinical assessments must be delivered in clinical settings by a trained clinician. Wearable technologies may be able to bring eye gaze and emotion recognition into natural social interactions and settings. Objective: This study aimed to test: (1) the feasibility of tracking gaze using wearable smart glasses during a facial expression recognition task and (2) the ability of these gaze-tracking data, together with facial expression recognition responses, to distinguish children with autism from neurotypical controls (NCs). Methods: We compared the eye gaze and emotion recognition patterns of 16 children with autism spectrum disorder (ASD) and 17 children without ASD via wearable smart glasses fitted with a custom eye tracker. Children identified static facial expressions of images presented on a computer screen along with nonsocial distractors while wearing Google Glass and the eye tracker. Faces were presented in three trials, during one of which children received feedback in the form of the correct classification. We employed hybrid human-labeling and computer vision–enabled methods for pupil tracking and world–gaze translation calibration. We analyzed the impact of gaze and emotion recognition features in a prediction task aiming to distinguish children with ASD from NC participants. Results: Gaze and emotion recognition patterns enabled the training of a classifier that distinguished ASD and NC groups. However, it was unable to significantly outperform other classifiers that used only age and gender features, suggesting that further work is necessary to disentangle these effects. Conclusions: Although wearable smart glasses show promise in identifying subtle differences in gaze tracking and emotion recognition patterns in children with and without ASD, the present form factor and data do not allow for these differences to be reliably exploited by machine learning systems. Resolving these challenges will be an important step toward continuous tracking of the ASD phenotype.
- Open Access
- 10 July 2019
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 21  (7) , e13094
Identification and Quantification of Gaps in Access to Autism Resources in the United States: An Infodemiological Study
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: 1 in 59 children have autism. While similar rates of autism are reported in rural and urban areas, rural families report greater difficulty in accessing resources. An overwhelming number of families experience waitlists for diagnostic and therapeutic services. Objective: To accurately identify gaps in access to autism care using GapMap: a mobile platform that connects families with local resources while continuously collecting up-to-date autism resource epidemiology information. Methods: Resources were extracted from various resource databases. Resources were deduplicated, validated, and allocated into seven categories based on keywords identified on the resource website. The average distance between individuals from a simulated autism population and the nearest autism resource in our database was calculated for each U.S. county. Resource load, an approximation of demand over supply for diagnostic resources, was calculated for each U.S. county. Results: There are approximately 28,000 U.S. resources validated on the GapMap database, each allocated into one or more of seven categories. States with the greatest distances to autism resources include Alaska, Nevada, Wyoming, Montana, and Arizona. Of the seven resource categories, diagnostic resources are the most underrepresented, comprising only 8.83% of all resources. Alarmingly, 83.86% of all U.S. counties lack any diagnostic resource. States with the highest diagnostic resource load include West Virginia, Kentucky, Maine, Mississippi, and New Mexico. Conclusions: Results from this study demonstrate the sparsity and uneven distribution of diagnostic resources in the U.S., which may contribute to the lengthy waitlists and travel distances often necessary to receive diagnosis in specific regions. More data is needed on autism diagnosis demand to better quantify resource needs across the U.S.
- 10 July 2019
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 21  (7) , e13094
Open AccessIdentification and Quantification of Gaps in Access to Autism Resources in the United States: An Infodemiological Study
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: 1 in 59 children have autism. While similar rates of autism are reported in rural and urban areas, rural families report greater difficulty in accessing resources. An overwhelming number of families experience waitlists for diagnostic and therapeutic services. Objective: To accurately identify gaps in access to autism care using GapMap: a mobile platform that connects families with local resources while continuously collecting up-to-date autism resource epidemiology information. Methods: Resources were extracted from various resource databases. Resources were deduplicated, validated, and allocated into seven categories based on keywords identified on the resource website. The average distance between individuals from a simulated autism population and the nearest autism resource in our database was calculated for each U.S. county. Resource load, an approximation of demand over supply for diagnostic resources, was calculated for each U.S. county. Results: There are approximately 28,000 U.S. resources validated on the GapMap database, each allocated into one or more of seven categories. States with the greatest distances to autism resources include Alaska, Nevada, Wyoming, Montana, and Arizona. Of the seven resource categories, diagnostic resources are the most underrepresented, comprising only 8.83% of all resources. Alarmingly, 83.86% of all U.S. counties lack any diagnostic resource. States with the highest diagnostic resource load include West Virginia, Kentucky, Maine, Mississippi, and New Mexico. Conclusions: Results from this study demonstrate the sparsity and uneven distribution of diagnostic resources in the U.S., which may contribute to the lengthy waitlists and travel distances often necessary to receive diagnosis in specific regions. More data is needed on autism diagnosis demand to better quantify resource needs across the U.S.
- Open Access
- 21 March 2019
- Published by JMIR Publications Inc. in JMIR Preprints
Toward Continuous Social Phenotyping: Analyzing Gaze Patterns in an Emotion Recognition Task for Children With Autism Through Wearable Smart Glasses (Preprint)
BACKGROUND Several studies have shown that facial attention differs in children with autism. Measuring eye gaze and emotion recognition in children with autism is challenging, as standard clinical assessments must be delivered in clinical settings by a trained clinician. Wearable technologies may be able to bring eye gaze and emotion recognition into natural social interactions and settings. OBJECTIVE This study aimed to test: (1) the feasibility of tracking gaze using wearable smart glasses during a facial expression recognition task and (2) the ability of these gaze-tracking data, together with facial expression recognition responses, to distinguish children with autism from neurotypical controls (NCs). METHODS We compared the eye gaze and emotion recognition patterns of 16 children with autism spectrum disorder (ASD) and 17 children without ASD via wearable smart glasses fitted with a custom eye tracker. Children identified static facial expressions of images presented on a computer screen along with nonsocial distractors while wearing Google Glass and the eye tracker. Faces were presented in three trials, during one of which children received feedback in the form of the correct classification. We employed hybrid human-labeling and computer vision–enabled methods for pupil tracking and world–gaze translation calibration. We analyzed the impact of gaze and emotion recognition features in a prediction task aiming to distinguish children with ASD from NC participants. RESULTS Gaze and emotion recognition patterns enabled the training of a classifier that distinguished ASD and NC groups. However, it was unable to significantly outperform other classifiers that used only age and gender features, suggesting that further work is necessary to disentangle these effects. CONCLUSIONS Although wearable smart glasses show promise in identifying subtle differences in gaze tracking and emotion recognition patterns in children with and without ASD, the present form factor and data do not allow for these differences to be reliably exploited by machine learning systems. Resolving these challenges will be an important step toward continuous tracking of the ASD phenotype.
- 21 March 2019
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessToward Continuous Social Phenotyping: Analyzing Gaze Patterns in an Emotion Recognition Task for Children With Autism Through Wearable Smart Glasses (Preprint)
BACKGROUND Several studies have shown that facial attention differs in children with autism. Measuring eye gaze and emotion recognition in children with autism is challenging, as standard clinical assessments must be delivered in clinical settings by a trained clinician. Wearable technologies may be able to bring eye gaze and emotion recognition into natural social interactions and settings. OBJECTIVE This study aimed to test: (1) the feasibility of tracking gaze using wearable smart glasses during a facial expression recognition task and (2) the ability of these gaze-tracking data, together with facial expression recognition responses, to distinguish children with autism from neurotypical controls (NCs). METHODS We compared the eye gaze and emotion recognition patterns of 16 children with autism spectrum disorder (ASD) and 17 children without ASD via wearable smart glasses fitted with a custom eye tracker. Children identified static facial expressions of images presented on a computer screen along with nonsocial distractors while wearing Google Glass and the eye tracker. Faces were presented in three trials, during one of which children received feedback in the form of the correct classification. We employed hybrid human-labeling and computer vision–enabled methods for pupil tracking and world–gaze translation calibration. We analyzed the impact of gaze and emotion recognition features in a prediction task aiming to distinguish children with ASD from NC participants. RESULTS Gaze and emotion recognition patterns enabled the training of a classifier that distinguished ASD and NC groups. However, it was unable to significantly outperform other classifiers that used only age and gender features, suggesting that further work is necessary to disentangle these effects. CONCLUSIONS Although wearable smart glasses show promise in identifying subtle differences in gaze tracking and emotion recognition patterns in children with and without ASD, the present form factor and data do not allow for these differences to be reliably exploited by machine learning systems. Resolving these challenges will be an important step toward continuous tracking of the ASD phenotype.
- Open Access
- 12 December 2018
- Published by JMIR Publications Inc. in JMIR Preprints
Identification and Quantification of Gaps in Access to Autism Resources in the United States: An Infodemiological Study (Preprint)
BACKGROUND Autism affects 1 in every 59 children in the United States, according to estimates from the Centers for Disease Control and Prevention’s Autism and Developmental Disabilities Monitoring Network in 2018. Although similar rates of autism are reported in rural and urban areas, rural families report greater difficulty in accessing resources. An overwhelming number of families experience long waitlists for diagnostic and therapeutic services. OBJECTIVE The objective of this study was to accurately identify gaps in access to autism care using GapMap, a mobile platform that connects families with local resources while continuously collecting up-to-date autism resource epidemiological information. METHODS After being extracted from various databases, resources were deduplicated, validated, and allocated into 7 categories based on the keywords identified on the resource website. The average distance between the individuals from a simulated autism population and the nearest autism resource in our database was calculated for each US county. Resource load, an approximation of demand over supply for diagnostic resources, was calculated for each US county. RESULTS There are approximately 28,000 US resources validated on the GapMap database, each allocated into 1 or more of the 7 categories. States with the greatest distances to autism resources included Alaska, Nevada, Wyoming, Montana, and Arizona. Of the 7 resource categories, diagnostic resources were the most underrepresented, comprising only 8.83% (2472/28,003) of all resources. Alarmingly, 83.86% (2635/3142) of all US counties lacked any diagnostic resources. States with the highest diagnostic resource load included West Virginia, Kentucky, Maine, Mississippi, and New Mexico. CONCLUSIONS Results from this study demonstrate the sparsity and uneven distribution of diagnostic resources in the United States, which may contribute to the lengthy waitlists and travel distances—barriers to be overcome to be able to receive diagnosis in specific regions. More data are needed on autism diagnosis demand to better quantify resource needs across the United States.
- 12 December 2018
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessIdentification and Quantification of Gaps in Access to Autism Resources in the United States: An Infodemiological Study (Preprint)
BACKGROUND Autism affects 1 in every 59 children in the United States, according to estimates from the Centers for Disease Control and Prevention’s Autism and Developmental Disabilities Monitoring Network in 2018. Although similar rates of autism are reported in rural and urban areas, rural families report greater difficulty in accessing resources. An overwhelming number of families experience long waitlists for diagnostic and therapeutic services. OBJECTIVE The objective of this study was to accurately identify gaps in access to autism care using GapMap, a mobile platform that connects families with local resources while continuously collecting up-to-date autism resource epidemiological information. METHODS After being extracted from various databases, resources were deduplicated, validated, and allocated into 7 categories based on the keywords identified on the resource website. The average distance between the individuals from a simulated autism population and the nearest autism resource in our database was calculated for each US county. Resource load, an approximation of demand over supply for diagnostic resources, was calculated for each US county. RESULTS There are approximately 28,000 US resources validated on the GapMap database, each allocated into 1 or more of the 7 categories. States with the greatest distances to autism resources included Alaska, Nevada, Wyoming, Montana, and Arizona. Of the 7 resource categories, diagnostic resources were the most underrepresented, comprising only 8.83% (2472/28,003) of all resources. Alarmingly, 83.86% (2635/3142) of all US counties lacked any diagnostic resources. States with the highest diagnostic resource load included West Virginia, Kentucky, Maine, Mississippi, and New Mexico. CONCLUSIONS Results from this study demonstrate the sparsity and uneven distribution of diagnostic resources in the United States, which may contribute to the lengthy waitlists and travel distances—barriers to be overcome to be able to receive diagnosis in specific regions. More data are needed on autism diagnosis demand to better quantify resource needs across the United States.
- Top Cited PapersOpen Access
- 27 November 2018
- Published by Public Library of Science (PLoS) in PLOS Medicine
- Vol. 15  (11) , e1002705
Mobile detection of autism through machine learning on home video: A development and prospective validation study
The standard approaches to diagnosing autism spectrum disorder (ASD) evaluate between 20 and 100 behaviors and take several hours to complete. This has in part contributed to long wait times for a diagnosis and subsequent delays in access to therapy. We hypothesize that the use of machine learning analysis on home video can speed the diagnosis without compromising accuracy. We have analyzed item-level records from 2 standard diagnostic instruments to construct machine learning classifiers optimized for sparsity, interpretability, and accuracy. In the present study, we prospectively test whether the features from these optimized models can be extracted by blinded nonexpert raters from 3-minute home videos of children with and without ASD to arrive at a rapid and accurate machine learning autism classification. We created a mobile web portal for video raters to assess 30 behavioral features (e.g., eye contact, social smile) that are used by 8 independent machine learning models for identifying ASD, each with >94% accuracy in cross-validation testing and subsequent independent validation from previous work. We then collected 116 short home videos of children with autism (mean age = 4 years 10 months, SD = 2 years 3 months) and 46 videos of typically developing children (mean age = 2 years 11 months, SD = 1 year 2 months). Three raters blind to the diagnosis independently measured each of the 30 features from the 8 models, with a median time to completion of 4 minutes. Although several models (consisting of alternating decision trees, support vector machine [SVM], logistic regression (LR), radial kernel, and linear SVM) performed well, a sparse 5-feature LR classifier (LR5) yielded the highest accuracy (area under the curve [AUC]: 92% [95% CI 88%–97%]) across all ages tested. We used a prospectively collected independent validation set of 66 videos (33 ASD and 33 non-ASD) and 3 independent rater measurements to validate the outcome, achieving lower but comparable accuracy (AUC: 89% [95% CI 81%–95%]). Finally, we applied LR to the 162-video-feature matrix to construct an 8-feature model, which achieved 0.93 AUC (95% CI 0.90–0.97) on the held-out test set and 0.86 on the validation set of 66 videos. Validation on children with an existing diagnosis limited the ability to generalize the performance to undiagnosed populations. These results support the hypothesis that feature tagging of home videos for machine learning classification of autism can yield accurate outcomes in short time frames, using mobile devices. Further work will be needed to confirm that this approach can accelerate autism diagnosis at scale.
- 27 November 2018
- Published by Public Library of Science (PLoS) in PLOS Medicine
- Vol. 15  (11) , e1002705
Top Cited PapersOpen AccessMobile detection of autism through machine learning on home video: A development and prospective validation study
The standard approaches to diagnosing autism spectrum disorder (ASD) evaluate between 20 and 100 behaviors and take several hours to complete. This has in part contributed to long wait times for a diagnosis and subsequent delays in access to therapy. We hypothesize that the use of machine learning analysis on home video can speed the diagnosis without compromising accuracy. We have analyzed item-level records from 2 standard diagnostic instruments to construct machine learning classifiers optimized for sparsity, interpretability, and accuracy. In the present study, we prospectively test whether the features from these optimized models can be extracted by blinded nonexpert raters from 3-minute home videos of children with and without ASD to arrive at a rapid and accurate machine learning autism classification. We created a mobile web portal for video raters to assess 30 behavioral features (e.g., eye contact, social smile) that are used by 8 independent machine learning models for identifying ASD, each with >94% accuracy in cross-validation testing and subsequent independent validation from previous work. We then collected 116 short home videos of children with autism (mean age = 4 years 10 months, SD = 2 years 3 months) and 46 videos of typically developing children (mean age = 2 years 11 months, SD = 1 year 2 months). Three raters blind to the diagnosis independently measured each of the 30 features from the 8 models, with a median time to completion of 4 minutes. Although several models (consisting of alternating decision trees, support vector machine [SVM], logistic regression (LR), radial kernel, and linear SVM) performed well, a sparse 5-feature LR classifier (LR5) yielded the highest accuracy (area under the curve [AUC]: 92% [95% CI 88%–97%]) across all ages tested. We used a prospectively collected independent validation set of 66 videos (33 ASD and 33 non-ASD) and 3 independent rater measurements to validate the outcome, achieving lower but comparable accuracy (AUC: 89% [95% CI 81%–95%]). Finally, we applied LR to the 162-video-feature matrix to construct an 8-feature model, which achieved 0.93 AUC (95% CI 0.90–0.97) on the held-out test set and 0.86 on the validation set of 66 videos. Validation on children with an existing diagnosis limited the ability to generalize the performance to undiagnosed populations. These results support the hypothesis that feature tagging of home videos for machine learning classification of autism can yield accurate outcomes in short time frames, using mobile devices. Further work will be needed to confirm that this approach can accelerate autism diagnosis at scale.
- Open Access
- 2 August 2018
- Published by Springer Nature in npj Digital Medicine
- Vol. 1  (1) , 1-10
Exploratory study examining the at-home feasibility of a wearable tool for social-affective learning in children with autism
Although standard behavioral interventions for autism spectrum disorder (ASD) are effective therapies for social deficits, they face criticism for being time-intensive and overdependent on specialists. Earlier starting age of therapy is a strong predictor of later success, but waitlists for therapies can be 18 months long. To address these complications, we developed Superpower Glass, a machine-learning-assisted software system that runs on Google Glass and an Android smartphone, designed for use during social interactions. This pilot exploratory study examines our prototype tool’s potential for social-affective learning for children with autism. We sent our tool home with 14 families and assessed changes from intake to conclusion through the Social Responsiveness Scale (SRS-2), a facial affect recognition task (EGG), and qualitative parent reports. A repeated-measures one-way ANOVA demonstrated a decrease in SRS-2 total scores by an average 7.14 points (F(1,13) = 33.20,p = <.001, higher scores indicate higher ASD severity). EGG scores also increased by an average 9.55 correct responses (F(1,10) = 11.89,p = <.01). Parents reported increased eye contact and greater social acuity. This feasibility study supports using mobile technologies for potential therapeutic purposes.
- 2 August 2018
- Published by Springer Nature in npj Digital Medicine
- Vol. 1  (1) , 1-10
Open AccessExploratory study examining the at-home feasibility of a wearable tool for social-affective learning in children with autism
Although standard behavioral interventions for autism spectrum disorder (ASD) are effective therapies for social deficits, they face criticism for being time-intensive and overdependent on specialists. Earlier starting age of therapy is a strong predictor of later success, but waitlists for therapies can be 18 months long. To address these complications, we developed Superpower Glass, a machine-learning-assisted software system that runs on Google Glass and an Android smartphone, designed for use during social interactions. This pilot exploratory study examines our prototype tool’s potential for social-affective learning for children with autism. We sent our tool home with 14 families and assessed changes from intake to conclusion through the Social Responsiveness Scale (SRS-2), a facial affect recognition task (EGG), and qualitative parent reports. A repeated-measures one-way ANOVA demonstrated a decrease in SRS-2 total scores by an average 7.14 points (F(1,13) = 33.20,p = <.001, higher scores indicate higher ASD severity). EGG scores also increased by an average 9.55 correct responses (F(1,10) = 11.89,p = <.01). Parents reported increased eye contact and greater social acuity. This feasibility study supports using mobile technologies for potential therapeutic purposes.
- Open Access
- 4 May 2017
- Published by JMIR Publications Inc. in JMIR Public Health and Surveillance
- Vol. 3  (2) , e27
GapMap: Enabling Comprehensive Autism Resource Epidemiology
A multidisciplinary journal that focuses on public health and technology, public health informatics, mass media campaigns, surveillance, and innovation in public health practice and research. Also dedicated to rapid open data sharing during epidemics.
- 4 May 2017
- Published by JMIR Publications Inc. in JMIR Public Health and Surveillance
- Vol. 3  (2) , e27
Open AccessGapMap: Enabling Comprehensive Autism Resource Epidemiology
A multidisciplinary journal that focuses on public health and technology, public health informatics, mass media campaigns, surveillance, and innovation in public health practice and research. Also dedicated to rapid open data sharing during epidemics.