The Knit Health technology is completely unique in the field of sleep health assessment. Knit takes an unprecedented approach to evaluating sleep by analyzing pediatric sleep patterns using artificial intelligence and computer vision technology. Understanding the science behind Knit reveals why it is the only in-home pediatric sleep assessment solution that can signal the presence of potential sleep issues and provide daily feedback necessary for improving long-term sleep health.
Knit uses cutting-edge computer vision technology and an advanced artificial intelligence neural network to track respiration and identify breathing disturbances, sleep stages and other primary sleep health metrics. The camera tracks the key skeletal features of a subject - head, neck and coccyx - even through thick blankets. With this understanding as context, the camera applies sophisticated signal processing and artificial intelligence to measure many of the same metrics as those tracked in a sleep lab – but without any patient contact. As a proxy for the actigraph in a sleep lab, which determines sleep state for Knit, our camera measures whole body movement. Instead of using respiratory inductance plethysmography (RIP) to measure breathing, attached to the body with two bands across the chest and abdomen, the Knit camera leverages computer vision technology to track abdominal and thoracic movements. Breathing is also tracked in a sleep lab by the nasal cannula (nasal flow), and while the Knit technology might not have the exact same level of accuracy, our studies have shown excellent signal correlation without the discomfort of body- worn sensors and potential sensor disconnection.
There are traditional PSG data channels that Knit does not measure, which include pulse-ox and EEG. However, Knit produces longitudinal data over an unrestricted timespan, providing robust directional indicators and nuanced insight into a patient’s normal and/or disordered sleep pa erns. We believe this baseline data will significantly increase the power of interpretation and diagnosis of a single night of broader sleep lab data.
Foundations in Science and Expert Analysis
The Knit technology is built on both decades of sleep science research and significant input from medical experts in the field of sleep science. In order to provide a better experience for patients and to gather a comprehensive understanding of their sleep, Knit developed camera technology that is unobtrusive to the patient’s life and passively gathers data over multiple nights. Knit researchers acquired expertise in state-of-the-art sleep science, analyzing both polysomnography and actigraphy, in order to develop technology that makes Knit the first preventative step for pediatric sleep issues. To develop this unique ability to accurately measure respiration without wearables, researchers scrutinized dozens of journals and publications, including numerous studies by the National Institutes of Health, the Journal of Clinical Neurophysiology and additional academic studies by respected scholars (cited at the end of this article).
Knit has a panel of expert advisors in the fields of sleep science, pediatric sleep medicine, neuroscience, and pulmonology who have guided the development of the technology and continue to advise ongoing technical development, as well as provide expert review of patient data, recommendations and sleep assessment reports. For qualifying breathing disturbances, Knit employs board-registered polysomnography technologists (RPSGT) and board-certified sleep doctors, the same accreditation required to score PSG metrics in a sleep lab. This parallel education ensures that the scoring mechanisms used by Knit are clinically similar to those used in a sleep lab. This esteemed group of experts are all passionate advocates of the importance of sleep medicine and share the belief that Knit’s technology could revolutionize the discipline.
Proven Results Over Hundreds of Studies
As a result of its accessible, foolproof and passive technology, Knit has one key advantage over traditional sleep science research methods – volume. The vast majority of sleep studies rely on a tiny fraction of the number of patients and do not have the consecutive nights that Knit has gathered. Studies by Knit have been performed on more than 175 pediatric patients, accumulating over 10,000 nights of data. Built on a formula developed by one of Knit’s sleep medicine experts, Knit sleep assessments collect multiple nights of data to provide a more comprehensive picture of an individual’s sleep health while simultaneously contributing to the accumulated knowledge in the field of sleep health science.
Additionally, the Knit technology has been benchmarked against other trusted in-home polysomnography equipment and has been found to reliably measure a respiration signal that strongly correlates to that of the respiratory inductance plethysmography (RIP). This data is critical in sleep assessments, as it is a strong indicating factor in numerous sleep disorders.
Knit Health for Your Patients
The foundation of well-established science underlying the Knit technology combined with extensive validation of real-world data ensures that providers can trust Knit to give them accurate, unbiased and unprecedented visibility into their patients’ sleep health. Knit’s ease of use and availability for pediatric patients makes it the obvious choice for assessing sleep health and identifying the risk of potential sleep issues.
Knit Sleep Expert Advisors
Dr. Logan Schneider, MD
Dr. Schneider is an internationally recognized expert in the assessment and treatment of sleep disorders in children and adults. He works closely with Knit to apply his breadth of experience from clinical and scientific training in Neurology at Johns Hopkins and Sleep Medicine at Stanford. He applies expert recommendations for Knit’s sleep reports and cutting-edge, academic research to guide Knit’s breathing detection algorithms. Dr. Schneider believes that Knit will allow big data and machine learning to play a bigger role in sleep assessment and treatment.
Dr. Michelle Jonelis, MD
Dr. Jonelis is Board Certified in Neurology and Sleep Medicine. She is a practicing clinician trained in pulmonology, neurology and Cognitive Behavioral Therapy at both Stanford University and the University of California, San Francisco. Dr. Jonelis works closely with Knit to provide guidance on recommendations within the Knit sleep reports, delivering a strong behavioral as well as neurological perspective. She is particularly interested in non-pharmacological management of sleep disorders and believes that Knit can be a tool to guide families to alternative solutions.
Data Security is Paramount
To put a camera in one’s home requires trust in the institution of origin to be careful and responsible with your data. Knit strives to validate and deserve this trust with three tenets of privacy:
Video clips are only accessed by sleep experts
The camera monitors sleep and captures video only of the sleep events that matter. The user has private access to these clips, which are only accessed by Knit’s sleep experts if a sleep issue is detected.
The view of the camera is limited
The camera has a narrow frame of vision that only captures the sleeping area on the bed, not other parts of the room, keeping unrelated happenings in the home outside of the scope of the camera.
We take security seriously
The authenticated live video stream runs only on the user’s secure, local network with sleep data processing on the camera, making it a closed loop system. AES encryption and Transport Layer Security (TLS) are used for data transfer and storage between Knit cameras and servers, backed by the security standards of Amazon Web Services.
ACADEMIC SOURCES: Luciane de Souza MSc, Ana Amélia Benedito-Silva PhD, Maria Laura Nogueira Pires PhD, Dalva Poyares MD, PhD, Sergio Tufik MD, PhD, Helena Maria Calil MD, PhD. Instrumentation and Methods | Sadeh A, Sharkey KM, Carskadon MA. Activity-based sleep-wake identification: A empirical test of methodological issues | Cole RJ, Kripke DF, Gruen W, Mullaney DJ, Gillin JC. Automatic sleep/wake identification from wrist activity | Jean-Louis G, Von Gizycki H, Zizi F, et al. Determination of sleep and wakefulness with the Actigraph Data Analysis Software (ADAS) | Rechtschaffen A, Kales A. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects | Avi Sadeh, Department of Psychology, Tel Aviv University, and Christine Acebo, E. P. Bradley Hospital/Brown University Medical School. The role of actigraphy in sleep medicine
Knit is not an FDA regulated device and does not replace a diagnosis by a medical professional. Knit helps people improve their sleep by providing context and risk-levels for potential sleep issues.