Skip to main content
Phase I - Artificially Intelligent Handheld Raman Spectrometer

Fast detection of biological pathogens and chemical agents in a field setting

E
Written by Eric Adolphe
Updated over 5 months ago

January 01, 2024

The objective of the Phase I SBIR was to augment a miniature, COTS Raman system with ML/AI, new signal processing techniques, and Surface-Enhanced Raman Scattering (SERS) to address sensitivity and noise challenges.

In Phase I, Forward Edge-AI also developed new signal processing and ML/AI techniques to enable the detection of trace chemicals and microbes. Next, Forward Edge-AI, augmented a miniature, Commercial-Off-The-Shelf (COTS) Raman spectrometer with ML/AI, new signal processing techniques, and SERS for quick stand-off identification of trace chemicals and microbes (e.g., COVID) in a field setting, thereby bringing laboratory grade science to the edge. In Phase I, we demonstrated the ability to detect 32 different strains of bacteria and viruses down (100 CFU and 100 PFU limits of detection).

Below is a spreadsheet containing the Raman spectra of viruses and bacteria collected:

Forward Edge-AI developed a new signal processing technique by converting the Raman Spectra into audio as shown in the image below. Click below to access project videos:

Forward Edge-AI designed a sample holder and probe that includes the SERS consumable. A glass test-tube is contained inside the metal outer case.

The second objective was to develop a method to continuously update a chemical/biological agent threat database. The Phase I study revealed a requirement for a dedicated, “subscription-based” library of Raman spectra because existing threat databases available are outdated. The research included a Blockchain and Distributed Ledger Technology to manage the governance of Raman Spectra and award crypto currency to researchers who contribute Raman spectra.

Federated Learning Marketplace of Raman Spectra

Key Capabilities Developed:

  • Handheld Android Device, networked for Federated Learning of Raman spectra

  • New deep learning model and signal processing techniques combined with SERS to achieve high resolution and high level of detection sensitivity (biological pathogens and trace chemicals) within the context of Raman

  • New model compression technique to on-board ML/AI onto Internet of Medical Things (IoMT) devices

  • (Common dataset, Blockchain/DLT subscription service, and incentive model to continuously update chem/bio threat libraries

Forward Edge-AI also conducted a COTS Technology Survey including an exhaustive technology survey to identify a suitable miniature. Forward Edge-AI identified an innovative new Raman spectrometer developed with $150 in parts. Forward Edge-AI is currently negotiating a license to manufacture the technology.

Another objective of the Phase I project was to address bandwidth and security challenges. Current technologies on the market could not support dense ML/AI algorithms needed for local training, and real-time identification of emerging threats.

The solution extends to a wide range of related use cases including: 1) Security - identifying/analyzing trace explosives, narcotics, chemical warfare agents, and other toxic chemicals, 2) Consumer – low cost capability to scan common goods for dangerous pathogens, pesticides and other chemicals, 3) Water – similar to fuel, testing water quality in expeditionary environments can be expensive requiring lab-based testing supply chains that include refrigeration, 4) Life Sciences - single cells drug interactions, disease diagnosis, and 5) Pharmaceutics - uniformity and component distribution.

Related Research:

Stakeholders

Download Logos

Award a Phase III Sole Source Contract

A Federal Agency may enter into a Phase III SBIR/STTR agreement at any time with a Phase I OR II Awardee. A subcontract to a Federally funded prime contract may be a Phase III award.

  1. Step 1 Requirements Document: Prepare a Statement of Work (SOW), Statement of Objectives (SOO), or Performance Work Statement (PWS), or use our automated tool to generate a document

  2. Step 2 Market Research: Use this page as your market research, or view a list of other eligible projects, then request a Rough Order Magnitude (ROM) from Forward Edge-AI

  3. Step 3 Funding: Performed by the government

  4. Step 4 Sole Source Justification: A Memorandum for the Record is required in lieu of a J&A or SSJ

  5. Step 5 Provide Requirements Package to Contracting Officer: Performed by the government

  6. Step 6 Solicitation: Performed by the government

  7. Step 7 Pre-Negotiation Memorandum: Use GSA CALC as a benchmark to determine fair and reasonableness of our ROM

  8. Step 8: Contract Award: Performed by the government

Language for Step 4 (Determination and Finding):

  • Artificial Intelligence (CNN/RNN), Android ATAK, Federated Learning Marketplace, Machine Learning, Microsoft Azure, signal processing (Wavelet and Fourier transformation), Raman Spectroscopy

  • Contested logistics (fuel and water quality testing), Chemical, Biological, Radiological, and Nuclear (CBRN) detection, explosives detection, Expeditionary Environments

List of Phase III contracts awarded so far

  • Come back soon

SBIR DATA RIGHTS:

Awarding Agency: US Air Force

Contract Number: FA864921P1006

Contractor Name: Forward Edge-AI, Inc.

Contractor Address: 10108 Carter Canyon, San Antonio, TX 78255

Expiration of SBIR Data: 14 April 2041

Protection Period: 20 years from award of contract on 14 April 2021

The Government's rights to use, modify, reproduce, release, perform, display, or disclose technical data or computer software marked with this legend are restricted during the period shown as provided in paragraph (b)(5) of the Rights In Other Than Commercial Technical Data and Computer Software–Small Business Innovation Research (SBIR) Program clause contained in the above identified contract. After the expiration date shown above, the Government has perpetual government purpose rights as provided in paragraph (b)(5) of that clause. Any reproduction of technical data, computer software, or portions thereof marked with this legend must also reproduce the markings.

© 2024 Forward Edge-AI, Inc. All rights reserved.

Did this answer your question?