By evaluation of the bone age, physicians can predict conditions that affect the growth of children. This is relevant for the reason that medical experts can take appropriate measures in time against diseases that delay or prevent physical growth and development. PANDA is intended to support an objective and standardized determination of the pediatric bone age based on digital X-rays.
Manual estimation of bone age suffers from a high degree of inter-rater variability. Consequently, this creates challenges in its clinical application, comparison between subjects, and even follow-ups. Further, extracting relevant information can be tedious and time-consuming. Our PANDA software provides a swift automated method to estimate bone age, monitor child growth and development, allowing you to spend your time where it is needed the most – With your patients.

Value Proposition

Manual estimation of bone age suffers from a high degree of inter-rater variability. Consequently, this creates challenges in its clinical application, comparison between subjects, and even follow-ups. Further, extracting relevant information can be tedious and time-consuming. Our PANDA software provides a swift automated method to estimate bone age, monitor child growth and development, allowing you to spend your time where it is needed the most – With your patients.

Narrative

A pediatrician suspects a patient to suffer from some sort of growth problem: nanism, delayed growth,gigantism. The pediatrician refers the patient to a radiologist and requests a bone-age estimate, or a bone-age status: normal/delayed/accelerated. In the classical way the radiologist uses an atlas to measure the bone age and together with the biological/chronological age and sex of the patient provides a bone-age status of the patient, a time consuming process that fatigues the radiologist. With PANDA this can be done automatically within seconds. PANDA receives radiographs from a PACS, analyzes the images and stores the analysis results in the PACS. As soon as PANDA receives the image, the analysis and the generation of reports are fully automated and entail no user interaction besides the user viewing the reports. The user can send images to PANDA via standardized DICOM commands or the file interface and receive reports over the same interface. In this sense, the user does not “operate” the device but simply reviews the reports and can accept or reject them. The IB Lab PANDA software does not provide a graphical user interface that uses the DICOM-viewers of existing workflows by attaching reports to the original study. All relevant images can be pre-analyzed by PANDA automatically by appropriate filter rules in the PACS. With this the original workflow is seamlessly augmented by PANDA.

Intended Use

IB Lab PANDA is a fully automated, radiological image processing software intended to aid medical professionals in the estimation of pediatric bone age according to the Greulich & Pyle method on hand radiographs of children aged between 36 months and 172 months (girls) or 192 months (boys). The software should not be used in-lieu of full patient evaluation or solely relied upon to make or confirm a diagnosis. The system is to be used by trained professionals including, but not limited to, radiologists, pediatricians and endocrinologists.

Limitations

IB Lab PANDA can be used for analysing PA digital  X-rays of the non-dominant hand only for children aged between 36 months and 172 months (girls) or 192 months (boys).The software should not be used in-lieu of full patient evaluation or solely relied upon to make or confirm a diagnosis. The system is to be used by trained professionals including, but not limited to, radiologists, pediatricians and endocrinologists.

Model performance metrics

PANDA is trained on a dataset consisting of 1384 DICOM left hand radiographs of normal children between 0 and 228 months of age, collected from the Children’s Hospital of Los Angeles. Each radiograph of the left hand was read for bone age according to the GP method by at least two pediatric radiologists.
PANDA performance in numbers:
4.36 months - mean absolute deviation (MAD) on the RSNA Pediatric Bone Age Machine Learning Challenge testing dataset.
6.2 seconds fully automated CPU-only analysis time (average value obtained on a 64-bit machine with i7-6700HQ CPU @ 2.60GHz x 8 and 16GB of RAM DDR4.)

ImageBiopsy Lab

ImageBiopsy Lab

PHILIP

Philip Meier

Email