SBIR-STTR Award

Neonatologists are often required to identify infants who are likely to suffer poor neurodevelopmental outcomes, including Cerebral Palsy (CP). CP is the most common motor disability among children in the United States and is associated with risk factors including low weight for gestational age, premature birth, and stroke. Although MRI and cranial ultrasound provide valuable structural information in the preterm period, they have moderate predictive accuracy for early CP risk identification. Over the past 20 years, numerous studies have validated the clinical potential of General Movement Assessment (GMA) for early CP risk identification and there is consensus in the literature that GMA offers the highest accuracy. Stage 1 "cramped synchronized" general movements (CSGMs) spanning 34-48 weeks gestational age (GA) during the "writhing movements" period and Stage 2 "forced, voluntary movements" spanning 50-59 weeks GA have demonstrated high sensitivity and specificity for developing CP, conjointly ranging from 80%-98% when performed by extensively trained experts. Despite its potential, GMA is available in very few clinical centers, as adoption and routine application depend on the availability of highly trained GMA raters to perform lengthy and costly bedside observations or video review- based scoring and manual report creation. A Cerebral Palsy Risk Identification System (CPRIS) is proposed that will be the first to automate GMA for routine application. The CPRIS constitutes a next-generation approach that will fundamentally transform GMA by replacing rater visual gestalts with objective, systematic, validated movement pattern classification. Further, the CPRIS potentially offers a means of informing, and assessing the efficacy of emerging stem cell-based interventions for CP along the early developmental continuum. Successful implementation of Phase I&II will complete a small form factor, mobile, highly automated preproduction system for cerebral palsy risk identification that can be readily applied by staff, clinicians, and health care provider personnel without any form of manual post-processing operations or video file transfer. An integrated utility will support GMA creation and report sharing with Electronic Health Record (EHR) systems. An application- specific, fully integrated device will achieve the highest degree of standardization and thus data quality. In a field study at two prominent Level 3 NICUs, infant movements will be acquired using an "RGB-D", or 3D "depth" camera in conjunction with an application- and stage-specific "Depth-Flow" convolutional neural network (CNN) classifier approach, that requires no infant contact (contrasting with kinematic methods) and captures whole- body movements. This effort marks the first utilization of such technology to automate GMA. Results will be compared to consensus determinations of advanced GMA raters in a sample of high risk preterm infants at both Stages 1 & 2. Viability of the new approach will be determined by ROC-AUC analyses, with a threshold for success of ≥ 0.90 accuracy. Overall results will be evaluated by an Advisory Committee of recognized experts in the fields of neonatology, pediatrics, cerebral palsy, GMA and biostatistics.

Public Health Relevance Statement:
PROJECT NARRATIVE Cerebral palsy is the most common physical disability in childhood. The overall project goal is to develop a computerized hardware-software system capable of identifying infants at high risk of developing cerebral palsy based on the systematic identification of specific patterns of movement-derived features. The Cerebral Palsy Risk Identification System (CPRIS) will enable clinical staff with no training in General Movement Assessment to collect data sets that will then be processed by a validated machine learning classifier. The CPRIS constitutes a key enabling technology for advancement in the risk assessment of CP at the earliest possible stage along the developmental continuum.

Project Terms:
Adoption; Biometry; Biometrics; Biostatistics; Birth; Parturition; Cerebral Palsy; Certification; Child; 0-11 years old; Child Youth; Children (0-21); youngster; Classification; Systematics; Elements; Environment; Gestational Age; Chronologic Fetal Maturity; Fetal Age; Goals; Health Personnel; Health Care Providers; Healthcare Providers; Healthcare worker; health care personnel; health care worker; health provider; health workforce; healthcare personnel; medical personnel; treatment provider; Infant; Premature Infant; infants born premature; infants born prematurely; premature baby; premature infant human; preterm baby; preterm infant; preterm infant human; Literature; Magnetic Resonance Imaging; MR Imaging; MR Tomography; MRI; MRIs; Medical Imaging, Magnetic Resonance / Nuclear Magnetic Resonance; NMR Imaging; NMR Tomography; Nuclear Magnetic Resonance Imaging; Zeugmatography; Manuals; Methods; Movement; body movement; Muscle Cramp; Cramp; Muscular Cramp; Neonatology; Out-patients; Outpatients; Pediatrics; Phenotype; Risk; Risk Factors; Scoring Method; Sensitivity and Specificity; Software; Computer software; Standardization; Progenitor Cells; stem cells; Apoplexy; Brain Vascular Accident; Cerebral Stroke; Cerebrovascular Apoplexy; Cerebrovascular Stroke; brain attack; cerebral vascular accident; cerebrovascular accident; Stroke; Technology; Testing; Time; United States; Video Recording; Videorecording; video recording system; Weight; physically handicapped; physical disability; physically disabled; Risk Assessment; Data Set; Dataset; Premature Birth; Prematurely delivering; Preterm Birth; premature childbirth; premature delivery; preterm delivery; Advisory Committees; Task Forces; advisory team; base; Cephalic; Cranial; Distal; Site; Clinical; Phase; Ensure; Training; disability; pediatric; Childhood; Visual; Data Bases; data base; Databases; data quality; Progress Reports; Sample Size; Multi-center studies; Multicenter Studies; Consensus; Complex; Pattern; System; 3-D; 3D; three dimensional; 3-Dimensional; Best Practice Analysis; Benchmarking; meetings; experience; field based data; field learning; field test; field study; Performance; success; kinematic model; kinematics; Devices; Manpower; personnel; Human Resources; Reporting; Position; Positioning Attribute; Sampling; Intervention Strategies; interventional strategy; Intervention; data processing; computerized data processing; Data; device development; instrument development; Device or Instrument Development; Motor; Process; Development; developmental; Electronic Health Record; electronic health care record; electronic healthcare record; cost; computerized; software systems; design; designing; next generation; Clinical assessments; novel strategies; new approaches; novel approaches; novel strategy; Outcome; clinical application; clinical applicability; inclusion criteria; high risk; operation; convolutional neural network; ConvNet; convolutional network; convolutional neural nets; neural network classifier; clinical center; wireless; machine learning classifier; machine learning based classifier; ultrasound

Neonatologists are often required to identify infants who are likely to suffer poor neurodevelopmental outcomes, including Cerebral Palsy (CP). CP is the most common motor disability among children in the United States and is associated with risk factors including low weight for gestational age, premature birth, and stroke. Although MRI and cranial ultrasound provide valuable structural information in the preterm period, they have moderate predictive accuracy for early CP risk identification. Over the past 20 years, numerous studies have validated the clinical potential of General Movement Assessment (GMA) for early CP risk identification and there is consensus in the literature that GMA offers the highest accuracy. Stage 1 "cramped synchronized" general movements (CSGMs) spanning 34-48 weeks gestational age (GA) during the "writhing movements" period and Stage 2 "forced, voluntary movements" spanning 50-59 weeks GA have demonstrated high sensitivity and specificity for developing CP, conjointly ranging from 80%-98% when performed by extensively trained experts. Despite its potential, GMA is available in very few clinical centers, as adoption and routine application depend on the availability of highly trained GMA raters to perform lengthy and costly bedside observations or video review- based scoring and manual report creation. A Cerebral Palsy Risk Identification System (CPRIS) is proposed that will be the first to automate GMA for routine application. The CPRIS constitutes a next-generation approach that will fundamentally transform GMA by replacing rater visual gestalts with objective, systematic, validated movement pattern classification. Further, the CPRIS potentially offers a means of informing, and assessing the efficacy of emerging stem cell-based interventions for CP along the early developmental continuum. Successful implementation of Phase I&II will complete a small form factor, mobile, highly automated preproduction system for cerebral palsy risk identification that can be readily applied by staff, clinicians, and health care provider personnel without any form of manual post-processing operations or video file transfer. An integrated utility will support GMA creation and report sharing with Electronic Health Record (EHR) systems. An application- specific, fully integrated device will achieve the highest degree of standardization and thus data quality. In a field study at two prominent Level 3 NICUs, infant movements will be acquired using an "RGB-D", or 3D "depth" camera in conjunction with an application- and stage-specific "Depth-Flow" convolutional neural network (CNN) classifier approach, that requires no infant contact (contrasting with kinematic methods) and captures whole- body movements. This effort marks the first utilization of such technology to automate GMA. Results will be compared to consensus determinations of advanced GMA raters in a sample of high risk preterm infants at both Stages 1 & 2. Viability of the new approach will be determined by ROC-AUC analyses, with a threshold for success of ≥ 0.90 accuracy. Overall results will be evaluated by an Advisory Committee of recognized experts in the fields of neonatology, pediatrics, cerebral palsy, GMA and biostatistics.

Public Health Relevance Statement:
PROJECT NARRATIVE Cerebral palsy is the most common physical disability in childhood. The overall project goal is to develop a computerized hardware-software system capable of identifying infants at high risk of developing cerebral palsy based on the systematic identification of specific patterns of movement-derived features. The Cerebral Palsy Risk Identification System (CPRIS) will enable clinical staff with no training in General Movement Assessment to collect data sets that will then be processed by a validated machine learning classifier. The CPRIS constitutes a key enabling technology for advancement in the risk assessment of CP at the earliest possible stage along the developmental continuum.

Project Terms:
Adoption; Biometry; Biometrics; Biostatistics; Birth Weight; Birth; Parturition; Cerebral Palsy; Certification; Child; 0-11 years old; Child Youth; Children (0-21); kids; youngster; Classification; Systematics; Elements; Environment; Gestational Age; Chronologic Fetal Maturity; Fetal Age; Goals; Health Personnel; Health Care Providers; Healthcare Providers; Healthcare worker; health care personnel; health care worker; health provider; health workforce; healthcare personnel; medical personnel; treatment provider; Infant; Premature Infant; infants born premature; infants born prematurely; premature baby; premature infant human; preterm baby; preterm infant; preterm infant human; Literature; Magnetic Resonance Imaging; MR Imaging; MR Tomography; MRI; MRIs; Medical Imaging, Magnetic Resonance / Nuclear Magnetic Resonance; NMR Imaging; NMR Tomography; Nuclear Magnetic Resonance Imaging; Zeugmatography; Manuals; Methods; Movement; body movement; Muscle Cramp; Cramp; Muscular Cramp; Neonatology; Outpatients; Out-patients; Pediatrics; Phenotype; Production; Risk; Risk Factors; Scoring Method; Sensitivity and Specificity; Software Validation; Software Verification; Standardization; stem cells; Progenitor Cells; Stroke; Apoplexy; Brain Vascular Accident; Cerebral Stroke; Cerebrovascular Apoplexy; Cerebrovascular Stroke; brain attack; cerebral vascular accident; cerebrovascular accident; stroked; strokes; Technology; Testing; Time; United States; Video Recording; Videorecording; video recording system; Weight; weights; physical disability; physically disabled; physically handicapped; Risk Assessment; Data Set; Prematurely delivering; Preterm Birth; premature childbirth; premature delivery; preterm delivery; Premature Birth; Task Forces; advisory team; Advisory Committees; Cranial; Cephalic; Distal; Site; Clinical; Phase; Ensure; Training; disability; pediatric; Childhood; Visual; Data Bases; data base; Databases; data quality; Progress Reports; Sample Size; Multi-center studies; Multicenter Studies; Consensus; Complex; Pattern; System; 3-Dimensional; 3-D; 3D; three dimensional; Benchmarking; Best Practice Analysis; benchmark; meetings; meeting; experience; field study; field based data; field learning; field test; Performance; success; kinematics; kinematic model; Devices; Human Resources; Manpower; personnel; Reporting; Positioning Attribute; Position; Sampling; Intervention; Intervention Strategies; interventional strategy; data processing; computerized data processing; Data; Device or Instrument Development; device development; instrument development; Motor; Process; Development; developmental; cost; computerized; software systems; designing; design; next generation; determine efficacy; efficacy analysis; efficacy assessment; efficacy determination; efficacy examination; evaluate efficacy; examine efficacy; efficacy evaluation; new approaches; novel approaches; novel strategy; novel strategies; Outcome; clinical applicability; clinical application; inclusion criteria; high risk; operations; operation; ConvNet; convolutional network; convolutional neural nets; convolutional neural network; neural network classifier; clinical center; wireless; machine learning classifier; machine learning based classifier; ultrasound; electronic health record system; EHR system