Juan Pablo Centeno, Senior Software Engineer II

Juan Pablo Centeno

SENIOR SOFTWARE ENGINEER II

Joined: April 10, 2023

Summary

Deep learning for imaging is my specialty: CT, X-ray, and otoscopy algorithms since 2017, an FDA cleared coronary calcium product, a RadUnity 510(k) amendment, and PhotoniCare's otitis media detector. At Innolitics I have contributed to 14 projects for 9 medical device clients since 2023 and led 7 of them as project manager, including PhotoniCare's otitis media detection algorithm from digital otoscopy and a real-time tympanic membrane segmentation MVP. Earlier, at National Jewish Health and Imbio, I wrote 510(k) documentation and co-authored a 2019 Radiology paper on deep learning classification of emphysema at CT and an International Forum of Allergy and Rhinology paper on automated sinus opacification quantification. I am currently supporting verification and validation for a De Novo AI/ML chest X-ray SaMD and cybersecurity documentation for a 510(k).

Skills

AI/ML Computer Vision Medical Images DICOM

Bio

I am an AI/ML Software Engineer specializing in the medical field. My expertise lies in developing and implementing AI/ML algorithms for medical device development and ensuring regulatory compliance, including FDA clearance. In my current role, I have developed a multi-organ mouse segmentation algorithm for CT images and its integration with PET scans, as well as an automatic tool for assessing brain aneurysm post-endovascular coiling from x-ray images. I have contributed to the creation of an algorithm for detecting cardiomegaly in CT scans and prototyped a system for the automatic detection of skull fractures in CT images.
During my previous roles, I successfully designed and implemented innovative solutions, including the automation of coronary artery calcium segmentation and quantification from chest CT scans through deep learning. I also developed advanced deep learning-based systems for the segmentation and quantification of paranasal sinus opacification from head CT scans, lung fibrosis quantification, and the segmentation of diseased lung tissues. My work included the development of algorithms for subject-level classification of emphysema and detection of interstitial lung diseases from chest CTs. My ongoing commitment is to enhance the capabilities of medical imaging, making it more precise and accessible for healthcare providers.

Hobbies

Triathlon, Ski