Sr. Glucose Algorithms Profusa
South San Francisco, CA

Job Description

Summary

Profusa is developing novel biosensors to continuously monitor tissue analytes in the body to provide unprecedented insights into human health physiology. We are searching for a professional with a strong background in continuous glucose monitoring (CGM) to lead the development and implementation of calibration and correction algorithms for our glucose-sensitive fluorescence-based subcutaneous microimplant.

You will be responsible for creating and executing methods to transform the transcutaneously-collected fluorescent signals into a real-time CGM data stream with high accuracy. You will lead the development of correction algorithms (for tissue variations, temperature, oxygen, etc.), and signal processing/filtering methods to optimize CGM system performance.

As a key member of the Sensor Development team, you will collaborate and communicate with other departments including software, firmware/hardware (optical engineering), and clinical to help design and verify optimal CGM platforms.

Essential Duties and Responsibilities

* Perform linear and nonlinear system identification of real-time fluorescent signals from subcutaneously implanted glucose sensors, and use signal processing, time-series analysis, mathematical modelling, and statistical analysis to determine appropriate calibration functions to convert fluorescence signal into analyte concentration with high accuracy.
* Apply data mining and machine learning techniques to extract relevant information and patterns from stochastic human physiological data and other measurements to develop correlation algorithms and calibration paradigms to improve system accuracy.
* Manage, organize, and analyze an influx of data generated from in-vitro bench tests, preclinical studies, and clinical trials to assess sensor performance. Fit data to biochemical and physical models, perform regression analysis, T tests, correlation analysis, etc. to refine algorithms.
* Design experiments to isolate and characterize error sources in sensor signals. Create programs to statistically analyze the impact of errors on performance, and quantitatively evaluate the efficacy of correction algorithms on reducing error (Bland-Altman, Clarke/Parkes EG, MARD).
* Characterize sensor dynamics and signal artifacts to develop mathematical models to simulate sensor behavior in implanted tissue environments and apply such models to drive optimization of correction algorithms and advanced sensor/hardware designs.
* Establish relationship between in-vitro and optimal in-vivo parameters to derive factory calibration parameters for real time estimation of subcutaneous glucose concentration. Design experiments to test the accuracy of a retrospectively-determined single calibration factor.

Education Requirements

Ph.D. in Computer Science, Engineering, Chemistry, Physics, or related fields with 4+ years of direct, relevant industry experience

Experience Requirements

* Experience collecting and analyzing clinical CGM data
* Proficiency in computing languages (Python, R, Matlab, etc.) is required
* Proficiency in programming languages (C/C++, Java, Javascript) is preferred
* Sound understanding of statistics, analog and digital signal processing techniques applied to biological data
* Experience in pattern recognition, and machine learning with physiological signal processing
* Excellent written and verbal communication skills with a history of scientific publications and oral presentations
* Systematic thinker who displays a blend of theoretical rigor and empirical approach to achieve practical, deadline driven task completion