AN EXPLORATION OF MOBILE APPLICATION FOR INFANT MOVEMENT SCREENING: PARENTAL PERCEPTION AND VIDEO RECORDING

Y.-C. Hsiao¹, Y.-J. Hsu², P.-N. Tsao³, T.-A. Yen³, W.-C. Liao⁴, W.-J Chen⁵, S.-F. Jeng¹

¹National Taiwan University, School and Graduate Institute of Physical Therapy, Taipei, Taiwan, ²National Taiwan University, Department of Computer Science & Information Engineering, Taipei, Taiwan, ³National Taiwan University Hospital, Department of Pediatrics, Taipei, Taiwan, ⁴National Taiwan University Hospital, Department of Internal Medicine, Taipei, Taiwan, ⁵National Taiwan University, Graduate Institute of Epidemiology and Preventive Medicine, Taipei, Taiwan

Background: Globally, 8.4% of children below 5 years of age have developmental disabilities. Early assessment of infants who are at risk of developmental disabilities like preterm infants is important for early intervention, particularly in remote areas. Artificial intelligence (AI) is increasingly applied for diagnosis of diseases on imaging findings but rarely for infant movement tracking and classification. To explore the application of AI for infant motor screening, our research team developed a mobile application (APP) “Baby Go” based on the Alberta Infants Motor Scale (AIMS) (58 items) to encourage parents to upload their infants’ movement videos from home. However, the accuracy of parental perception of infant movements and the quality of home videos remains unclear.

Purpose: To examine whether parents accurately perceive their infant’s movements and record good-quality of home videos via the APP “Baby Go.”

Methods: This study included 10 preterm infants and 16 term infants at 4 months of age. Parents were instructed the function of “Baby Go” that contains an education module regarding how to shoot good-quality home videos. The AIMS items with photos and feature descriptions were illustrated for infant movements at each age. After video recording, the parents were asked to find the movement items best fitting the features and then uploaded the videos via the APP during 4 to 18 months. A physiotherapist reviewed and classified the movement videos according to the AIMS criteria and then assessed the quality of videos. The accuracy of parental perception was defined when the parent’s selection of movement items agreed with the results of the physiotherapist. The quality of home videos was assessed for the completeness of infant’s body parts in the view, angle of view, and camera movement. Good-quality was defined as the infant being covered one body part at most, angle of shooting allowing recognition of the movement, and stable camera movement.

Results: The parents have uploaded 475 home videos when the infants were at 4 to 18 months of age. The mean percentage of correctly perceived movement items by the parents was 59.2% for the prone subscale, 73.7% for the supine subscale, 49.2% for the sitting subscale, and 58.5% for the standing subscale. Some sitting items were incorrectly perceived by parents because of the difficulty in understanding subtle weight-shifting movements and various position change maneuvers. The mean percentage of good-quality videos was 42.3% for the prone subscale, 57.1 % for the supine subscale, 42.5% for the sitting subscale, and 51.6% for the standing subscale. The common problems were shaky videos and incomplete body parts.

Conclusions: The parents were accurate in classifying infant’s movements in most of the AIMS subscales. Demonstration of videos and highlighting of crucial features of movements are warranted to enhance parental perception of infant movements. Meanwhile, strengthening of the education module is necessary to improve the quality of home videos.

Implications: The findings provide insightful information regarding parental perception of infant movements and quality of home videos to help enhance usefulness of the APP “Baby Go” for future development of artificial intelligence.

Funding acknowledgements: This study was funded by National science and technology council (MOST 110-2314-B-002-055-MY3).

Keywords:
Parental perception
Home videos
Infant movement screening

Topics:
Paediatrics
Innovative technology: information management, big data and artificial intelligence

Did this work require ethics approval? Yes

Institution: National Taiwan University Hospital

Committee: Research Ethics Committee

Ethics number: 202012089INB