ARTIFICIAL
INTELLIGENCE (AI)
SOLUTIONSCROLL DOWN
ABOUT
OUR COMPANY
Funnels Inc. promotes medical innovation by combining medical big data and AI. Our products will enable more accurate and faster diagnosis.
WHAT ARE WE
WORKING ON?
DEVELOP
Development of fast and
accurate diagnostic models
IMPROVE
Improving model accuracy
through continuous updates
EFFICIENT
Very few space,
time and cost constraints
MAJOR
REQUIREMENTS
SKILLS
AI experts with more than 10 years of experience developed the AI models
EXPANDABILITY
Possible expansion and development into related medical fields such as evaluation of scoliosis progression or prediction of functional recovery
ADAPTABILITY
No separate installation equipment required, available with just a computer and a web browser
FOCUS
Meeting the needs of users by combining medical big data with AI
EXPERTS
DEVELOP
THE HIGHEST
QUALITY AI
ALGORITHMS
Funnels Inc. promotes medical innovation by combining medical big data and AI. Our products will enable more accurate and faster diagnosis
BONE AGE
PREDICTION
ALGORITHM
BACKGROUND
The predicted growth height of children and adolescents is one of the great concerns of parents
Growth plates are closely related to bone age, and through bone age tests, the degree of bone maturity can be known, and future growth potential can be predicted
BONE AGE MEASUREMENT
Bone age and growth height can be predicted through X-ray of the hand, current height, and parent’s height
NOW
Diagnosis of bone age through
X-ray of the hand
FUTURE
Improve accuracy by measuring bone age in the elbow joint and pelvic bones
Combine images and clinical data
2,000
Pre-secured bone age diagnostic data
0.5
Target mean absolute error
14.83 years
Age prediction to two decimal places
DYSPHAGIA
DIAGNOSIS
ALGORITHM
BACKGROUND
Despite being able to objectively observe the entire process of swallowing through videofluoroscopic swallowing study (VFSS), its interpretation is complex and needs consideration of several factors
The application of the recent developments in deep learning research could reduce the burden over clinicians caused by the complexity of VFSS interpretation
DETECTING
PENETRATION OR ASPIRATION
Automatically detect penetration or aspiration in VFSS of patients with dysphagia
GOAL 1
Learning by extracting five consecutive frame images from VFSS video when the hyoid bone reaches its highest or lowest point during the swallowing process
GOAL 2
An algorithm for VFSS video analysis is being developed
SAMPLE IMAGE
950 for high-peak images and 950 for low-peak images
1,900
ACCURACY
Validation accuracy of deep learning models
94.7%
AUC
Area under the receiver operating characteristic curve
0.961
INTELLECTUAL PROPERTY STATUS
| Date of patent application | Title of patent | Patent application number |
|---|---|---|
| 2020-12 | Motor function prediction apparatus and method for determining need of ankle-foot-arthosis of stroke patients. | 10-2020-0171386 |
| 2020-12 | Apparatus and method for predicting recovery prognosis of motor function using clinical data from stroke patients. | 10-2020-0175177 |
| 2020-12 | Apparatus and method for predicting recovery prognosis of motor function using MRI image of corona radiata infarct. | 10-2020-0175263 |
| 2021-02 | Anterior cruciate ligament tear diagnbosis apparatus using machine hearning and method thereof. | 10-2021-0024513 |
| 2021-06 | Apparatus for determining presence of penetration or aspiration of dysphagia patient using VFSS and method thereof. | 10-2021-0075441 |
| 2021-17 | Apparatus for determining presence of cervical spondylotic myelopathy using radiographic images and method thereof. | 10-2021-0092130 |
| 2022-02 | Apparatus for determining meniscus tear using MRI images and method thereof. | 10-2022-0022166 |
| 2022-05 | Apparatus for determining osteonecrosis of the femoral head using X-ray image and method thereof. | 10-2022-0056396 |
| 2022-05 | Apparatus for determining cervical pseudoarthrosis and method thereof. | 10-2022-0063817 |
| 2022-07 | Apparatus for predicting treatment outcome of TFESI and method thereof. | 10-2022-0093958 |
| 2022-08 | System for measuring rehabilitation status of stroke patient based on mirror therapy and method thereof. | 10-2022-0105264 |
| 2022-12 | Apparatus and method for prescription of insoles in patient with foot pain using deep learning. | 10-2022-0184749 |
| 2023-01 | System for measuring rehabilitation status of stroke patient based on mirror therapy and method thereof. | 10-2023-0002233 |
OUR LOCATION
Yeungnam University Medical Center,
170, Hyeonchung-ro, Nam-gu, Daegu 42415,
Republic of Korea
CREDITS : This presentation template was created by Slidesgo, and includes icons by Flaticon, and infographics & images by Freepik.