UNITED STATES
SECURITIES AND EXCHANGE COMMISSION
Washington, D.C. 20549
FORM
10-K
(Mark One)
☒ANNUAL REPORT PURSUANT TO
SECTION 13 OR 15(d) OF THE SECURITIES EXCHANGE ACT OF 1934
For the fiscal year ended
December 31, 2023
Or
☐TRANSITION REPORT PURSUANT
TO SECTION 13 OR 15(d) OF THE SECURITIES EXCHANGE ACT OF 1934
For the transition period
from to
To
Commission File No. 001-40058
SPECTRAL
AI, INC.
(Exact name of registrant as specified
in its charter)
(Address of Principal Executive Offices) (Zip Code)
(972)499-4934
(Registrant’s telephone number, including area
code)
Securities registered pursuant to
Section 12(b) of the Act:
Title of each class Trading Symbol(s) Name of each exchange on which registered
Common stock, par value $0.0001 per share MDAI The Nasdaq Stock Market LLC
Securities registered
pursuant to Section 12(g) of the Act:
None
Indicate by check mark if the registrant
is a well-known seasoned issuer, as defined in Rule 405 of the Securities Act. Yes ☐ No ☒
Indicate by check mark if the registrant
is not required to file reports pursuant to Section 13 or Section 15(d) of the Act. Yes ☐ No ☒
Indicate by check mark whether the
registrant (1) has filed all reports required to be filed by Section 13 or 15(d) of the Securities Exchange Act of 1934 during the preceding
12 months (or for such shorter period that the registrant was required to file such reports), and (2) has been subject to such filing
requirements for the past 90 days. Yes ☒ No ☐
Indicate by check mark whether the
registrant has submitted electronically every Interactive Data File required to be submitted pursuant to Rule 405 of Regulation S-T (§
232.405 of this chapter) during the preceding 12 months (or for such shorter period that the registrant was required to submit such files).
Yes ☒ No ☐
Indicate by check mark whether the
registrant is a large accelerated filer, an accelerated filer, a non-accelerated filer, a smaller reporting company, or an emerging growth
company. See the definitions of “large accelerated filer,” “accelerated filer,” “smaller reporting company,”
and “emerging growth company” in Rule 12b-2 of the Exchange Act.
Large accelerated filer ☐ Accelerated filer ☐
Non-accelerated filer ☒ Smaller reporting company ☒
Emerging growth company ☒
If an emerging growth company, indicate
by check mark if the registrant has elected not to use the extended transition period for complying with any new or revised financial
accounting standards provided pursuant to Section 13(a) of the Exchange Act. ☐
Indicate by check mark whether the
registrant has filed a report on and attestation to its management’s assessment of the effectiveness of its internal control over
financial reporting under Section 404(b) of the Sarbanes-Oxley Act (15 U.S.C. 7262(b)) by the registered public accounting firm that prepared
or issued its audit report. ☐
If securities are registered pursuant
to Section 12(b) of the Act, indicate by check mark whether the financial statements of the registrant included in the filing reflect
the correction of an error to previously issued financial statements. ☐
Indicate by check mark whether any
of those error corrections are restatements that required a recovery analysis of incentive-based compensation received by any of the registrant’s
executive officers during the relevant recovery period pursuant to §240.10D-1(b). ☐
Indicate by check mark whether the registrant
is a shell company (as defined in Rule 12b-2 of the Exchange Act): Yes ☐ No ☒
The aggregate market value of the Registrant’s
Common Stock outstanding held by non-affiliates of the Registrant, computed as of September 12, 2023 (the date of completion of the registrant’s
Business Combination (as defined below)) was approximately $81,000,000.
As of March 25, 2024, there were 17,466,871 shares of Common Stock,
$0.0001 par value per share, issued and outstanding.
SPECTRAL AI, INC.
FORM 10-K FOR THE YEAR ENDED DECEMBER 31, 2023
TABLE OF CONTENTS
Page
Part I.
Item 1. Business 1
Item 1.A. Risk Factors. 11
Item 1.B. Unresolved Staff Comments. 48
Item 1.C. Cybersecurity 48
Item 2. Properties 49
Item 3. Legal Proceedings 49
Item 4. Mine Safety Disclosures 49
Part II.
Item 6. [Reserved] 51
Item 7.A. Quantitative and Qualitative Disclosures about Market Risk 60
Item 8. Financial Statements and Supplementary Data 60
Item 9.A. Controls and Procedures. 60
Item 9.B. Other Information. 61
Item 9.C. Disclosure Regarding Foreign Jurisdictions that Prevent Inspection. 61
Part III.
Item 10. Directors, Executive Officers and Corporate Governance 62
Item 11. Executive Compensation 62
Item 14. Principal Accountant Fees and Services 62
Part IV.
Item 15. Exhibits, Financial Statement Schedules F-1
i
CAUTIONARY NOTE REGARDING FORWARD-LOOKING
STATEMENTS AND RISK FACTOR SUMMARY
This Annual Report on Form 10-K
contains statements that are forward-looking and as such are not historical facts. This includes, without limitation, statements under
“Item 7. Management’s Discussion and Analysis of Financial Condition and Results of Operations” regarding our financial
position, business strategy and the plans and objectives of management for future operations. These statements constitute projections,
forecasts and forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. The words “anticipate,”
“believe,” “continue,” “could,” “estimate,” “expect,” “intends,”
“may,” “might,” “plan,” “possible,” “potential,” “predict,” “project,”
“should,” “will,” “would” and similar expressions may identify forward-looking statements, but the
absence of these words does not mean that a statement is not forward-looking.
The forward-looking statements contained
in this Annual Report on Form 10-K are based on our current expectations and beliefs concerning future developments and their potential
effects on us. There can be no assurance that future developments affecting us will be those that we have anticipated. These forward-looking
statements involve a number of risks, uncertainties (some of which are beyond our control) or other assumptions that may cause actual
results or performance to be materially different from those expressed or implied by these forward-looking statements. These risks and
uncertainties include, but are not limited to, the following risks, uncertainties and other factors:
ii
● The price of our Common Stock and Warrants may be volatile.
Should one or more of these risks
or uncertainties materialize, or should any of our assumptions prove incorrect, actual results may vary in material respects from those
projected in these forward-looking statements. We undertake no obligation to update or revise any forward-looking statements, whether
as a result of new information, future events or otherwise, except as may be required under applicable securities laws.
iii
PART I.
References in this Annual Report
on Form 10-K (this “Annual Report”) to “we,” “us,” “our” or the “Company”
are to Spectral AI, Inc., a Delaware corporation. References to our “management” or our “management team” refer
to our officers and directors.
Item
1. Business. Overview
We are an AI company focused
on predictive medical diagnostics. Our DeepView System uses proprietary AI algorithms to distinguish between fully damaged, partially
damaged and healthy human tissue characters invisible to the naked eye, at the initial time point of wound presentation. The DeepView
System delivers a binary prediction on the wounds capacity to heal or not-heal by a specified time point in the future. Our DeepView System’s
output is specifically engineered to assist the physician in making a more accurate, timely and informed decision regarding the treatment
of the patient’s wounds. Our focus from 2013 through 2021 was on the burn indication. In 2022, we expanded our focus to include
the DFU indication.
We were notified that our DeepView
System, comprised of the multispectral imaging (“MSI”) component integrated with the predictive AI-Burn® software
component received United Kingdom Conformity Assessed (UKCA) marking for use in the United Kingdom for burn indications on February 22,
2024. The UKCA marking registration was fully completed on March 7, 2024. We anticipate that our full DeepView System may achieve Class
II medical device designation in the with the United States Food and Drug Administration (FDA) via a De Novo application. Subject to our
receipt of additional necessary market authorization, our business will have two revenue streams, a SaaS model component predicated on
utilizing the regulatory method, SaMD (software as a medical device), and the imaging device component. The SaaS component will feature
a software licensing fee that includes maintenance, image hosting, and access to algorithm updates. The proprietary imaging device acquires
the images for the AI algorithms and is a universal platform to house multiple clinical applications including burn and DFU. Pricing
for these components will be evaluated and strategically set per country and site-of-service for heightened customer adoption.
The MSI imaging technology,
which comprises one part of the DeepView System, consists of patented proprietary multi-spectral optics and sensors, capturing injured
tissue images ranging from near UV lights, through the human visible wavelengths, all the way into the near infrared range (NIR). The
broad wavelength ranges go beyond what the human eyes can see and capture what medical professionals cannot observe with their naked eyes.
This wide range of wavelength images contains wound tissue physiology and captures the viability of various biomarkers within the skin
and from the injured tissue spectral signatures. The imaging technology extracts appropriate clinical data, processes the image data to
provide the injured tissue spectral signatures to the AI model and algorithms. The AI algorithm classifies various severities of the injuries
as (i) full damaged (non-healing), (ii) partially damaged or (iii) healthy tissue (healing) and displays a comparison of
the original image next to an image with a color overlay of the non-healing portions of the wound. The image acquisition takes 0.2 seconds,
and all image processing and AI model classification takes approximately 20 to 25 seconds. Our DeepView System’s proprietary optics
can extract millions of pixels of data or AI model features from each group of raw images. This information is then used to build and
continually improve the AI model, which is trained and tested against a proprietary and clinically validated database of approximately
340 billion pixels of DFU and burn data as of December 31, 2023. The DeepView-AI Burns® software is used with
the DeepView SnapShot® imaging device, and it is intended to be used as an adjunctive tool to aid health care providers
in the assessment of burn wound healing potential by differentiating non-healing from healing burned tissue within an image.
Below at Figure 1 is an example
of the DeepView System technological process.
Figure 1 — DeepView Imaging technology
1
To our knowledge, there are
no digital wound healing assessment in predictive medical diagnostic products that provide clinicians with an objective and immediate
assessment of a wound’s future healing potential and that benefit from the application of AI. Currently, healthcare professionals
rely on their experience and subjective assessments to determine if wounds, such as burn injuries and DFUs, will heal under routine care
after a period of time, typically several weeks, or are in need of advanced wound care products and procedures including surgical interventions.
Our DeepView System allows health care professionals to make a “Day One” assessment of a wound’s healing potential over
time.
We have received substantial support from the U.S. government
for our DeepView System’s application for burn wounds, including from agencies such as BARDA, which is part of the HHS Office of
the Assistant Secretary for Preparedness and Response (“ASPR”) in the United States, established to aid in securing the
United States from chemical, biological, radiological, and nuclear threats, as well as from pandemic influenza and emerging infectious
diseases. We have also received funding from the National Science Foundation (“NSF”), National Institute of Health (“NIH”)
and the DHA an agency within the Department of Defense (“DoD”). Since 2013, we have been awarded approximately $279.6 million
in funding from government contracts, substantially all of which is from BARDA, which accounts for $272.9 million. This has allowed
us to develop our technology and further our clinical trials. On September 27, 2023, the Company executed a new contract with BARDA,
providing the Company with additional funding of up to $150.0 million, including an initial award of approximately $54.9 million
to support the clinical validation and application for FDA De Novo status of our DeepView AI – Burn software. This will include
the distribution of up to 30 DeepView Systems in various emergency rooms and burn centers to support the clinical validation study. The
contract also includes options, similar to our prior BARDA contracts, with an additional total value of approximately $95.1 million
which can be exercised for additional product development, procurement and the expanded deployment of DeepView Systems at emergency rooms,
trauma and burn centers. These deployments will enable the Company to conduct health economic and outcome research to support the broader
clinical adoption of the DeepView System. This grant funding is non-dilutive to our stockholders, and we believe it validates the important
nature of our mission and technology.
Subject to our receipt of the
necessary regulatory market authorizations, we intend to initially sell the DeepView System throughout the United States and the
UK for its burn and its DFU indication. Given our receipt of the UKCA authorization for our burn indication we anticipate initial sales
in UK in the second half of 2024. The sales channel for these two indications are different. We expect that our burn indication will be
supported by existing and future governmental contracts, primarily from agencies such as BARDA and the DHA, while the DFU indication will
be an add-on to the burn indication sales channel and will have its own separate sales channel to penetrate the podiatric and wound care
clinics. In the United States, there are approximately 100 burn centers, 700 trauma centers and 5,400 federal and community hospitals
with Emergency Rooms where the burn patients are most likely to visit upon injuries. The DeepView System provides a quick clinical decision
tool to the emergency rooms, so it can be decided quickly whether patients need routine care or should be transferred to trauma centers
or burn centers for advanced care, and for quick and accurate surgical planning. In the burn centers, the DeepView System provides an
advanced guidance on the non-healing areas of a burns, Therefore, we plan to target our sales efforts to these facilities through our
highly-trained technical sales support staff that we plan to hire given the nature of DeepView as a truly disruptive AI driven predictive
assistance tool. For the DeepView System’s burn application and following receipt of any future contract awards, we plan to partner
with the U.S. governmental agency sponsors to implement the distribution of our DeepView System throughout the United States
into key regions to support the United Stats’ mass casualty countermeasure directives, with the goal of making our country better
prepared for mass casualty events and saving scarce healthcare resources.
Subject to our receipt of the
necessary market authorizations, we plan to begin our commercial sales efforts of the DeepView System’s DFU application in the UK
through key clinical sites and related networks. We expect to engage contract sales organizations to distribute our DeepView System throughout
the UK as well as eventually in the Netherlands, Germany, Italy and Spain (“EU4”). Preliminary discussions with distributors
occurred during 2023 to determine which organizations possess the key relationships and insights for selling diagnostic systems within
their respective countries. We intend to focus our commercial strategy initially in the UK, which we are targeting for mid-2024, with
the EU4 to follow in 2026, subject to CE mark approval for our technologies. Similar to the United States, the primary customer base
for the DFU application in Europe will be outpatient wound centers and secondary sites of care that have a high-volume of DFU patients.
We also expect to engage internal and/or third party resources to help us navigate the various regional tender and contracting entities
within each country. In the United States, subject to our receipt of the necessary regulatory market authorization, we anticipate
initially distributing the DeepView Systems using our DFU indication in hospitals’ emergency rooms and trauma centers. We will then
build in additional indications, given that we can run multiple indications on the same imaging devices. In addition, wound care centers
are typically the first line of specialty care for DFUs in the United States. Vascular and cardiology companies and outpatient podiatry
practices also treat wounds. We will need to grow our distribution network to support the expanded sales efforts for the DFU indication
to these facilities by initially focusing on management companies that have multiple podiatric and/or wound care centers under their management.
In this way, we believe we can build a mature sales model, pricing structure, and customer instructions, to enable us to further grow
our distribution networks with third-parties and other sales channel sources.
2
As noted above, subject to
our receipt of the necessary regulatory market authorizations, our business is expected to have two revenue streams, a SaaS model component
predicated on utilizing the regulatory method, SaMD (software as a medical device), and an imaging device component. The SaaS component
will feature a software licensing fee that includes maintenance, image hosting, and access to algorithm updates. The capital sale component
will be competitively priced for acceptance into independent practices and clinics.
In 2018, the FDA designated
our earlier version of the DeepView System with BDD status for its burn indication. The FDA’s designation as a Breakthrough Device
(“BDD”) allows for prioritized reviews and a dedicated line of communication with reviewing members of the FDA. In the
first quarter of 2021, the Health Products Regulatory Authority of Ireland (HPRA) provided a medical device classification recommendation
of IIa for our DeepView System. We have enrolled subjects in our DFU studies in clinical and academic sites across the United States
and Ireland. In 2022, we completed our first DFU clinical training study with 100 adult subjects in the United States at five
well-known medical facilities. In the third quarter of 2022, we extended the AI training study with an additional 100 adult subjects.
We completed this study in January 2023, providing us with a much-improved DFU AI prediction performance at 86%. In April 2023,
we commenced our validation study with an additional 200 adult subjects at 10 well-known medical facilities. This study is expected to
be completed in 2024. We have also signed with international partners, including well-respected institutions in the field and have partnered
with leading wound care physicians. We believe that we will be able to leverage these relationships to access other institutions and individuals,
which should increase awareness and early adoption of our technology in the United States, the UK and the EU. U.S. adoption
will also benefit from the potential future BARDA funding of technology placement for burns applications. Our focus will be on the continued
development of the DFU AI model as we progress through the validation study.
We expect to complete the validation
studies for the DFU regulatory application in the United States in 2024, while targeting for the FDA’s grant of our De Novo
petition in early 2025.
Subject to our receipt of the
necessary regulatory market authorization, we would expect to leverage results from the U.S. study for a simultaneous conformity
assessment procedure in the EU to obtain the CE marking of conformity (“CE Mark”), and we would expect to commence post-market
studies in the UK and Germany. Subject to our receipt of the necessary regulatory market authorization, we would expect to initiate commercialization
in the United States during 2025 and intend to submit for FDA review of the burn application in 2025 in accordance with the projected
timeline for our BARDA contract.
Burn Indication
We began conducting our validation
study for burn in the early 2024, where we plan to enroll an additional 240 adult and pediatric subjects at up to 20 clinical sites.
In adult participants, the
DeepView GEN3 System has shown 92% accuracy, with cross-validation from the AI model for identification of non-healing burn regions.
This represents a significant improvement above the diagnostic accuracy of burn physicians assessing the same adult burn patients, and
above 50% to 75% accuracy, according to industry literature. In addition, in head-to-head clinical trial evaluations, our DeepView System
provided higher accuracy on “Day One” to “ground truth” determined on day 21 on burn wound analysis than the accuracy
of burn specialists, reporting at 70 – 80% accuracy, and non-burn specialist physicians, reporting at 50 – 60%
accuracy. We have conducted three large clinical studies with multiple sites across the United States, enrolling 413 patients, including
329 adult burn patients and 84 pediatric patients. Through these studies we were able to determine burn assessment accuracy in both healing
and non-healing wounds.
In pediatric patients, the
AI performance of the DeepView System showed 88% accuracy, underlining how the AI technology is responding with significant reliability
to variability in the study population. Based on these strong results, we have bolstered our infrastructure to facilitate the expansion
of the study to additional sites and have begun enrollment in a larger study in order to complete the AI algorithm’s development.
As of December 31, 2023,
our proprietary and clinically validated database for burns is comprised of approximately 340 billion pixels of DFU and burn data. This
database presents both a significant barrier to entry to would-be competitors in wound care healing assessment, and a potential additional
commercial opportunity for us to develop further in the future.
3
DFU Indication
We made substantial progress
in our U.S. DFU Clinical Validation Study (the “US DFU Clinical Study”) in 2023. The endpoint that we are pursuing in
the clinical study is to predict on “Day One” whether the DFU wound will reduce in size by 50% by week four. Our DeepView
System showed improvement of the AI diagnostic accuracy to 86%.
The data collected from the
US DFU Clinical Study will be used to augment our existing proprietary and clinically validated database of DFU data and healthcare matrix
information; and to validate the DeepView DFU AI algorithm as we prepare for U.S. regulatory submission in 2024.
In the first half of 2023,
we continued to enroll subjects in the US DFU Clinical Study to finalize our admission goal. Additionally, we increased investment in
the DFU indication in 2023 to drive our commercialization strategy. We intend to submit for U.K. Conformity Assessment (“UKCA”)
regulatory evaluations in mid-2024. We are currently targeting to receive the required UKCA certificates in 2024, and to receive FDA marketing
authorization in 2025, although these authorizations cannot be guaranteed, and may take longer than expected.
In February 2023, we also
initiated a clinical study in the EU with the Royal College of Surgeons in Ireland conducted at Connolly Hospital in Dublin, Ireland.
The EU clinical study will collect data from DFU patients monitored for up to 12 weeks. The intention of the clinical study is to
further develop the DeepView AI algorithm to support our regulatory submissions for UKCA, FDA, and EU CE Mark. The imaging system that
makes up our DeepView System recently received United Kingdom Conformity Assessed (UKCA) marking for use in the United Kingdom and has
Class I medical device classification with the United States Food and Drug Administration (FDA), while we anticipate that the
DeepView System as a whole, including the AI component, may achieve Class II classification in the US via a De Novo application.
Other DeepView Programs in Development
Funding from the U.S. government
has also allowed us to develop additional “Horizon” indication uses of our DeepView System, including DeepView Snapshot M,
DeepView AI 3-D wound measurement technology, and other indications. We believe that our DeepView System’s use in emergency rooms,
trauma and burn centers and other would care facilities should be expanded to provide greater utility of the DeepView System in such settings.
DeepView SnapShot M
In addition to our DeepView
System, our primary additional technology is the DeepView SnapShot® M, a fully handheld, portable, wireless diagnostic
tool based on the DeepView System’s AI platform. The DeepView SnapShot M provides a potential enhanced and expanded use for the
U.S. government and emergency care, first responders and potentially home health care professionals. On June 23, 2021, we were
awarded a two-year, $1.1 million, Sequential Phase II STTR contract by the DHA within the U.S. Department of Defense. This
funding enables us to research and develop the DeepView SnapShot M product primarily for military and combat settings. In April 2023,
we were awarded a $4.0 million grant from the Medical Technology Enterprise Consortium (“MTEC”), a 501I(3) biomedical
technology consortium working in partnership with the Department of Defense, to develop our DeepView SnapShot M device in a Phase III
feasibility and commercialization study. This grant was increased by over $500,000 on March 12, 2024. These grants, along with prior awards
from DHA, bring our funding total for our DeepView SnapShot® M to over $6.0 million. The funding will be used to support
military battlefield burn evaluation using DeepView SnapShot M.
3-D Wound Measurement Technology
We are also currently developing
3-D software-based wound measurement technology for our DeepView System. This technology will produce rapid, accurate and easy-to-use
wound size measurement images to produce an accurate 3-D tissue representation from a single image snapshot enabling distance, area and
volume measurements with sub-millimetric accuracy without reference to any attendant markers or manually placed stickers or multiple images.
We believe this is a significant improvement over current wound size measurement technologies which are limited in their ability to measure
all three wound dimensions (distance, area and volume) or are otherwise cumbersome, requiring reference markers/stickers or multiple images
to determine would size measurements. Our 3D wound measurement technology calculates the total body surface area (“TBSA”)
of a wound. This technology will be integrated into our DeepView System and applies the “rule of nines”; a method that divides
the body’s surface area into percentages to calculate the size of a burn or wound. For example, the front and back of the head and
neck equal 9% of the body’s surface area and the front and back of each arm and hand equal 9% of the body’s surface area.
This technology enhancement will not only generate the TBSA measurement, but will also indicate the “healthy” versus “unhealthy”
tissue for advanced treatment applications to be applied to the burn or wound area. This is a critical step in assuring that these alternative
medical solutions will be successful in-patient applications. The 3-D wound size measurement tool has completed the proof-of-concept phase.
We are currently developing this technology in cooperation with BARDA.
Business Focus and Milestones
Our current focus is categorized
in two parts: (1) we will continue to fulfill our contractual obligations and meet milestones under our BARDA PBS contract (described
in further detail below); and (2) we will pursue the commercialization of the DFU application in the UK, United States and EU4.
Our near-term goals related to the BARDA PBS contract are to deliver on the current phase of the contract (Phase 1a), and to complete
the remaining phases of the BARDA PBS contract. Completion of these contractual phases support our long-term goal of entering into a federal
procurement contract with BARDA.
4
We intend to submit a De Novo
application to the FDA for market authorization of the burn application in early 2025. In 2023, we received our ISO 13485:2016 certification
for Medical Devices. The certification audit is expected to occur in the first quarter of 2024. In parallel, we are in the process of
scheduling the DeepView System Technical Documentation audit necessary to obtain the CE Mark and UKCA certificates to allow market access
in the EU and UK, respectively. In March, Spectral completed its UKCA Mark registration for the full DeepView System for our burn indication.
Figure 2 below provides a summary of our key anticipated regulatory submissions. There can be no assurance that we will be able to obtain
market authorization in the US, UK or EU of our DeepView GEN 3 System with AI on our projected timeline, or at all.
Figure 2 — Summary
of key regulatory submissions
DeepView in Practice
DeepView is a predictive analytics
platform that combines AI algorithms and MSI imaging for an assessment of wound healing potential. It is non-invasive, non-radiation,
non-laser and does not require the use of injectable dye. This integration can be characterized into four distinct components: DeepView
imaging, data extraction, AI model building and AI wound healing potential assessment. The DeepView AI®- Burn software
is used with the DeepView SnapShot® imaging device, and it is intended to be used as an adjunctive tool to aid health care
providers in the assessment of burn wound healing potential by differentiating non-healing from healing burned tissue within an image.
5
Figure 3 — Illustration of DeepView’s
binary decision assist output where the colored region marks the predicted non-healing portion of the wound.
The DeepView System is designed
to assist clinicians in making accurate, timely, and informed decisions regarding the treatment of the patient’s wound. In the case
of DFUs, an assessment that the wound will not heal over time would provide the health care professional with the appropriate justification
to use an advanced wound care therapy on “Day One” as opposed to waiting 30 days using standard of care and potentially
losing the patient to follow-up or risking patient non-compliance with standard wound therapy. The current clinical accuracy of DeepView
in ongoing clinical trials is 86% for DFUs compared with current physician accuracy as low as 50%. Subject to FDA market authorization
of the product, for burn wounds, the clinician could make an immediate and objective determination for appropriate candidates for surgery
as well as determining what specific areas of the burn wound will require skin grafting. In ongoing clinical trials, DeepView’s
current accuracy for burn wounds non-healing predictions is 92%, compared with current physician accuracy of 50 to 75%, according to industry
literature. In addition, in head-to-head clinical trial evaluations, when compared to “ground truth” our DeepView System provided
higher accuracy of burn wound analysis than the accuracy of burn specialists, reporting at 70 – 80% accuracy, and
non-burn specialist physicians, reporting at 50 – 60% accuracy, when comparing “Day One” analysis to
“ground truth” determined on day 21. We have conducted three large clinical studies with multiple sites across the United States,
enrolling 413 patients, including 329 adult burn patients and 84 pediatric burn patients. Through these studies, we were able to
determine burn assessment accuracy in both surgery and non-surgical treatment.
See the table below for an
analysis of the current DeepView System’s benefits to patient care:
Burn DFU
Current Time to Decision 21 Days 30 Days
DeepView® Time to Decision Day 1 Day 1
Current Clinical Accuracy 50 – 75% 50%
DeepView® Accuracy in Ongoing Clinical Trials 92% 86%
DeepView® Estimated Cost savings ~$24,000 per stay ~$63,100 per stay
6
Key Strengths
We believe the following key
strengths will help us to maintain and grow our business going forward:
Market Leading Technology
We have developed proprietary
AI algorithms and imaging technology to assist clinicians to make more accurate and efficient treatment decisions in managing patient’s
wounds. This technology is the result of 13 years of research and development, thousands of hours of user feedback, and most
importantly, the continual commitment to ensuring that the output from DeepView answers a clinical question that is to meaningful physicians.
We own and control the entirety of our data pipeline. We only rely images and data that the DeepView System collects in a controlled clinical
environment and do not rely on stock images or databases for our algorithms. All optical technology has been developed in-house and is
specifically engineered to collect this imaging data. A current image of our cart-based DeepView System appears below in Figure 8.
Figure 4 — DeepView Generation 3
System
Unmet Clinical Need
The biggest unmet need for
clinicians treating DFU and burn wounds is the lack of a diagnostic tool that provides an objective wound healing determination on “Day
One.” While burns and DFUs appear to be very different types of wounds, they are in fact similar from the perspective of assessment
and diagnosis. The treatment pathways for each of these wounds can be generally characterized by a subjective initial assessment from
the physician followed by multiple weeks of clinical observation to assess whether or not the wound responded to treatment. Both
are primarily staged by their penetration depth into the skin and involvement of tissues below the skin in severe cases. Both DFUs and
burns are diagnosed by expert clinical opinion without the aid of objective diagnostic tools that provide a wound healing prediction.
Furthermore, the current methods of diagnosis rely on a “wait and see” approach that result in prolonged hospital stays and
costly delays in the delivery of definitive treatment. Our goal is to eliminate these costly delays between initial screening and the
delivery of a definitive treatment through the use of AI algorithms applied to our proprietary multispectral wound images.
Significant Market Opportunity
Geography— DeepView
has the potential to service a large total addressable market. We estimate that there are over 57,000 sites of clinical care in which
the technology could be placed in the United States and over 20,000 sites across the UK and EU4. For all geographies, these sites
include both acute inpatient hospitals and outpatient sites of care, in order to include physician offices. As we expand from the United States
into the UK and EU4, we will consider follow-on markets for commercial expansion, including the Middle East, among others.
Pipeline Applications— Though
we are currently focused on the DFU and burn applications for DeepView, there are other pipeline applications that we are considering
for future commercialization. As noted above, we have already received U.S. government funding for the development of our DeepView
SnapShot® M fully handheld device for use in combat, military and home health care uses. In connection with our BARDA contract,
we are working on expanding the indication usage of the DeepView System to incorporate a wound and burn measurement tool for clinicians. We
have also explored the technology’s potential for the assessment of critical limb ischemia, level of lower limb amputation selection,
post-operative perfusion assessment for peripheral interventions, and military applications. For all future pipeline applications, we
believe that the technology would remain constant, in that we will leverage our data analytics algorithms to improve predictive analyses.
With any new application, we would need to conduct one or more clinical studies to collect enough patient data to appropriately support
algorithm development for each new application. These new algorithms could easily be uploaded to existing machines in the future. From
a regulatory perspective, we believe that these follow-on applications would all follow a 510(k) clearance process although in some
cases, we may need to follow the De Novo classification or premarket approval pathway if we are not able to identify a predicate, or if
use of the device for a new indication is classified as a Class III device.
7
Existing and future revenue base from long
term U.S. Government Contracts — BARDA
On September 27,
2023, the Company executed a new contract with BARDA, providing the Company with additional funding of up to $150.0 million,
including an initial award of approximately $54.9 million to support the clinical validation and application for FDA De Novo
status of our DeepView System. This will include the distribution of up to 30 DeepView Systems in various emergency rooms and burn
centers to support the clinical validation study. The contract also includes options, similar to our prior BARDA contracts, with an
additional total value of approximately $95.1 million which can be exercised for additional product development, procurement
and the expanded deployment of DeepView Systems at emergency rooms, trauma and burn centers. These deployments will enable the
Company to conduct health economic and outcome research to support the broader clinical adoption of the DeepView System. This grant
funding is non-dilutive to our stockholders, and we believe it validates the important nature of our mission and technology.
Significant Wound Data Repository from Artificial
Neural Network
As of December 31, 2023,
approximately 340 billion pixels of proprietary DFU and burn data have been acquired and utilized for the deep learning algorithms
training. This presents a significant barrier to entry to would-be competitors in wound care healing assessments. The data collection
to clinical output, the flow, quality and control of the data pipeline is managed entirely by us. Our DeepView System uses deep learning
on its wound data repository to recognize patterns and correlations of injured tissue spectral signatures to produce reliable and reasonable
assessment for clinicians to make accurate and faster treatment decisions. We believe that our strategic partnerships with various leading
medical institutions and healthcare providers in the United States and Europe will enable us to access high quality image data and
build the world’s leading wound biopsy tissue database. Our AI algorithms are designed and trained to the clinical “ground
truth” that has been verified and vetted by various U.S. government agencies and leading clinicians in their respective fields.
They have not yet been reviewed or cleared by FDA.
Strategic Partnerships
We have developed strategic
partnerships with multiple clinical and academic partners. In the United States, we are currently engaged with leading research hospitals
that are enrolling subjects for our Burn AI training study. In the EU and UK, we have partnered with the Royal College of Surgeons Ireland,
as well as key opinion leaders to provide us with greater knowledge in the wound care sector. Our partnerships with these institutions
provide us with the opportunity to collaborate with leading wound care providers to develop effective early stage wound assessment technology.
We utilize these strategic partnerships to support the ongoing clinical validation studies we are using to develop our algorithmic model.
Each of our clinical study/trials include certain protocol requirements to ensure a uniform testing process for our technology.
Proven Experienced Management Team
Our board of directors and
senior management team have significant experience in the technology and healthcare sectors, with a track record of successful entrepreneurship,
operational acumen, strategic relationships and the ability to understand and navigate the complexities of healthcare. Our directors also
bring significant expertise from previous public company experience along with financial, governance and technical oversight.
Respected Advisory Board
We have established an Advisory
Board composed of industry experts and opinion leaders that will raise our profile. Its members provide us with external, industry-specific
perspectives and technical support.
Competition
To our knowledge, no other
predictive wound-healing diagnostic imaging technology is available to clinicians who treat wounds. DeepView’s competitive advantage
is that it is the only AI-enabled wound imaging technology that translates raw physiological data/images into an output that is directly
correlated to predictive wound healing.
Several companies have developed
wound imaging systems for burn injuries and DFUs; however, these systems incorporate technology such as spatial frequency domain imaging,
thermal imaging, photographic documentation, hyperspectral imaging, and near-infrared imaging that provide physiologic data to the physician.
Ultimately, this physiologic data appears to only provide an indirect linkage to wound healing and does not display a binary result of
“healing vs. non-healing.” Furthermore, the majority of systems in the wound care space are merely documentation tools that
record measurements of the wound for health record purposes and still rely upon subjective clinician opinion for treatment decisions.
The advent of a novel technology such as the DeepView System not only has the potential to disrupt the therapeutic pathway within the
wound care market, but also to create a new diagnostic market for wound care that did not exist previously for clinics and physicians,
subject to successful development of the device and FDA marketing authorization. As noted above, although our previous DeepView Systems
received 510(k) clearance, and we have received FDA BDD clearance for our DeepView GEN 3 System, there can be no assurance that we will
be able to obtain market authorization in the US, UK or EU.
Commercialization and Revenue Strategy
We intend to pursue the complete
development of our DeepView System and, if marketing authorization is obtained, to commercialize it on our own, or potentially with a
partner, in the United States and other regions. We currently have no sales, marketing or commercial product distribution capabilities and
have no experience as a company commercializing products. However, if necessary, we intend to hire appropriately to build the necessary
infrastructure and capabilities over time for the United States, and potentially other regions, following further advancement of
our DeepView System.
8
United States
Subject to our receipt of the
necessary regulatory marketing authorization, we intend to market our DeepView System using internal and third party resources to inpatient
and outpatient sites of care throughout the United States. Wound Care Centers are typically the first line of specialty care for
DFUs in the United States, but vascular and cardiology groups and outpatient podiatry practices also treat wounds. Sales will initially
target wound care centers and podiatry practices presiding in areas with high prevalence of diabetes such as the south and southeastern
areas of the United States. As noted above, subject to our receipt of the necessary regulatory marketing authorization, our business
is expected to have two revenue streams, a SaaS model component predicated on utilizing the regulatory method, SaMD (software as a medical
device), and an imaging device component. The SaaS component will feature a software licensing fee that includes maintenance, image hosting,
and access to algorithm updates. The capital sale component will be competitively priced for acceptance into independent practices and
clinics.
Given our recent receipt of
the UKCA mark for our burn indication, commercial sales are expected to commence in 2024 for the burn indication in the UK and in 2025
for DFU indication. In the United States, the Company will continue to perform under its new BARDA contract with respect to the burn indication
and will receive significant governmental funding prior to seeking FDA clearance of the DeepView System in late 2025.
Reimbursement
We expect to utilize our post-market
clinical evidence and health economic impact analysis to submit to NHS for reimbursement for its Burn indication in the United Kingdom.
Upon more market penetration, we will apply for NICE certification. In the United States, we expect the DeepView System will be used
in both inpatient and outpatient sites of service. The process of reimbursement varies greatly between the two. The DeepView burn indication
will be used both in EDs and Burn Centers. As clinical evidence is developed and utilization increases over the next several years, we
plan to apply for CPTÒ codes.
Adoption
We view our DeepView technology
as disruptive by nature and there will be those who will be slow to adopt it. This emphasizes the importance of having the right strategic
partnerships, institutions, and physician key opinion leaders as early adopters. We plan to engage in relationships that can act as key
opinion leaders to share their experience on why they adopted the DeepView technology. The adoption will be supported by a team of field
clinical educators and digital marketing campaigns.
Manufacturing Arrangements
We currently outsource all
of our manufacturing to a Contract Manufacturer. Cobalt Product Solutions (“Cobalt”), located in Plano, Texas, is involved
with manufacturing the current generation DeepView System and we anticipate that they will continue to do so for the foreseeable future.
In addition to Cobalt, we partner
with several other highly specialized contract manufacturers in the areas of optics, technology design, and electronics. We employ experienced
regulatory and quality control personnel to ensure that our manufacturing processes and quality management systems are in compliance with
FDA and EU regulations and standards. As we expand into the European market, we will most likely consider manufacturing devices in the
EU in preparation for commercialization. We do not have any plans to develop our own manufacturing facility at this time.
Intellectual Property
We strive to protect and enhance
the proprietary technologies that we believe are important to our business by seeking patents to cover our technology. We also rely on
trade secrets to protect aspects of our business that are not amenable to, or that we do not consider appropriate for, patent protection.
Our technology is protected
with issued and/or allowed patents across nine families of active patents:
● Burn/Wound Classification on MSI and PPG;
● Tissue classification on MSI and PPG;
● Amputation site analysis on MSI, ML and healthcare matrix;
9
● High-precision, multi-aperture, MSI snapshot imaging;
● Wound assessment based on MSI;
● Burn/histology assessment based on MSI and ML;
● High-precision, single-aperture MSI snapshot imaging; and
● Topological characterization and assessment of tissues using MSI and ML
We have 10 issued and allowed
U.S. patents with 5 U.S. patent applications pending. We have 10 issued and allowed international patents with 29 foreign and
international patent applications pending.
In addition, we support the
development of our brand and product offerings through trademark protection at the United States Patent and Trademark Office. As
of December 31, 2023, we maintain a portfolio of 64 trademarks and nine trademark applications pending relating to our DeepView System
product offerings. Our trademarks and pending trademark applications are spread over nine jurisdictions mostly in the UK the EU and China. It
is our intention to maintain these registrations indefinitely and to expand the number of jurisdictions in which we have registered trademarks
as deemed necessary to protect our freedom to use the marks and/or block competitors in additional markets. We will continue to look to
protect our intellectual property in the United States, UK and the EU as those are the first commercial markets for our products
and rely on third party experts to assist in doing this.
Facilities
Our corporate headquarters
is located in Dallas, Texas, where we occupy approximately 11,000 square feet of space under a lease agreement. The lease agreement for
our corporate headquarters expires in in December 2024 with an additional three monthly extensions through March 31, 2025.
Human Capital Resources and Employees
We employ a growing and highly
skilled employee base, including our sales force, and promote a culture of innovation to continuously iterate and enhance our products,
systems and commercial footprint. Our human capital objectives include, as applicable, identifying, recruiting, retaining, incentivizing
and integrating our existing and additional employees.
We anticipate the expansion
of our business in 2024 as we continue to build a focused and highly skilled team. At December 31, 2023 had 78 full-time employees
in the United States and UK. In 2024, we anticipate new hires will be made in all areas, in particular in operations, sales, marketing,
and government contracts. This will further enable us to meet our technology, IP, clinical, regulatory, and commercial goals in 2024
and beyond.
We have designed and implemented
our cash and stock compensation programs to attract, motivate, and retain our employees. We regularly review our compensation structure
to ensure that we remain competitive, reward top performance, and ensure internal equity, while maintaining proper fiscal governance.
Our compensation packages are designed based on market benchmarks. We offer robust benefits package including health (medical, dental
and vision) insurance, paid time off, paid parental leave, a retirement plan and life and disability coverage.
Business Combination
On September 11, 2023, we consummated the business combination (the
“Business Combination”) contemplated by that certain Business Combination Agreement, dated as of April 11, 2023, by and among
Rosecliff Acquisition Corp I (“RCLF”), Ghost Merger Sub I, Inc. (“Merger Sub I”), Ghost Merger Sub II, LLC (“Merger
Sub II”), and Spectral MD Holdings, Ltd. (“Legacy Spectral”), whereby Merger Sub I merged with and into Legacy Spectral
(the “First Merger”), with Legacy Spectral surviving the First Merger as a wholly owned subsidiary of RCLF and RCLF changed
its name to “Spectral AI, Inc.”, and, immediately following the First Merger, Legacy Spectral merged with and into Merger
Sub II (the “Second Merger”), with Merger Sub II surviving the Second Merger as a wholly owned subsidiary of RCLF (collectively,
the “Merger”). For additional information, please refer to our final prospectus, as amended, on Form S-4, filed with the SEC
on August 10, 2023.
Available Information
Our
internet address is www.spectral-ai.com. Our website and the information contained therein or linked thereto are not part of this Annual
Report. We make available free of charge through our internet website our annual reports on Form 10-K, quarterly reports on Form 10-Q,
current reports on Form 8-K, proxy statements, registration statements and amendments to those reports filed or furnished pursuant to
the Exchange Act as soon as reasonably practicable after we electronically file such material with, or furnish them to the SEC. The SEC
maintains a website that contains reports, proxy statements and other information regarding issuers that file electronically with the
SEC. These materials may be obtained electronically by accessing the SEC’s website at www.sec.gov.
10
Item 1.A. Risk Factors.
Investing in our securities
involves risks. Before you make a decision to buy our securities, in addition to the risks and uncertainties discussed above under “Cautionary
Note Regarding Forward-Looking Statements,” you should carefully consider the specific risks set forth herein. If any of these
risks actually occur, it may materially harm our business, financial condition, liquidity and results of operations. As a result, the
market price of our securities could decline, and you could lose all or part of your investment. Additionally, the risks and uncertainties
described in this prospectus or any prospectus supplement are not the only risks and uncertainties that we face. We may face additional
risks and uncertainties that are not presently known to us, or that we currently deem immaterial, which may also impair our business,
prospects, financial condition or operating results. The following discussion should be read in conjunction with our financial statements
and the financial statements of the Company and notes to the financial statements included herein.
Risks Related to Our Financial Condition and
Capital Requirements
We have incurred significant losses since
inception and may not be able to achieve significant revenues or profitability.
We have incurred substantial net losses since our inception. For the year
ended December 31, 2023 and the year ended December 31 2022, on a consolidated basis, we incurred a net loss of $20.9 million
and $2.9 million, respectively, and on a consolidated basis our cash balance at December 31, 2023 was $4.8 million. We had an
accumulated deficit of $32.8 million as of December 31, 2023. Our losses have resulted primarily from costs incurred in connection