Partner with the UCSF Hypoxia Lab

The UCSF Hypoxia Lab works with a range of collaborators including research teams from UCSF and international universities, device manufacturers, product designers, leading global health authorities, regulatory bodies and more. Together, we conduct the highest quality research to help improve device safety and equity.

We have conducted studies for countless devices that have received regulatory clearances and made immeasurable public health impact. We have an unparalleled track record of more than 10,000 study participants and extensive experience working with regulatory bodies.

If you are interested in conducting a study with the UCSF Hypoxia Lab in San Francisco or at our partner lab in Uganda, please reach out to our clinical research coordination team. Please include the following emails in your inquiry: hypoxialab@ucsf.edu.

Information about setting up a study, our study protocols, and data collection processes can be found here.

Device Testing Services & Studies

The Hypoxia Lab has developed numerous study protocols to test a variety of physiologic conditions relevant to medical devices as well as wearable technologies for home use. These include devices that measure blood oxygen saturation, blood pressure, cerebral oxygen saturation, hemoglobin concentration, carbon monoxide levels, met-hemoglobin levels, cerebral blood flow, respiratory rate, heart rate and more. Our protocols are designed to generate data suitable for submission to the US Food and Drug Administration (FDA) for device clearances and for conformity with the International Organization for Standardization (ISO). All studies are done with the institutional review board (IRB) approval of the UCSF Committee on Human Research. Studies conducted collaboratively with our partner lab in Uganda also have local IRB approval from the Uganda Heart Institute and Uganda National Council for Science and Technology.

Standard Pulse Oximeter Study in San Francisco

We determine the accuracy of pulse oximeters by exposing volunteers to short periods of low oxygen levels (75-100% SaO2) while monitoring how the oximeter responds. This helps us understand if these devices are accurate for everyone regardless of skin color during these low oxygen conditions.

Standard Pulse Oximeter Study in Uganda

Motion Testing

During pulse oximetry testing, we can include tools that mimic patient movements. These tools, which come from the manufacturer, are called motion fixtures. They help us simulate scenarios where a patient might be moving, allowing us to see how well the pulse oximeters perform in situations with motion during low oxygen conditions.

Carboxyhemoglobin Study

We assess pulse oximeters tailored for detecting carboxyhemoglobin (COHb), a hemoglobin type not carrying oxygen as usual. To ensure if pulse oximeters reliably detect elevated COHb levels during low-oxygen conditions, volunteers undergo brief low-oxygen exposure (75-100% SaO2) while we observe oximeter responses.

Methemoglobin Study

We assess pulse oximeters tailored for detecting methemoglobin (metHb), a hemoglobin type not carrying oxygen as usual. To ensure if pulse oximeters reliably detect elevated metHb levels during low-oxygen conditions, volunteers undergo brief low-oxygen exposure (75-100% SaO2) while we observe oximeter responses.

Low Perfusion Testing

With low perfusion testing, we can simulate low perfusion (reduced blood flow) using clamps or cuffs. We can also test in various body positions, like the Trendelenburg position. This allows us to examine how well devices perform under low oxygen conditions (75-100% SaO2) where blood flow may be limited or changed.

Cerebral Oximetry Study

We’re studying cerebral oximeters to gauge their accuracy in measuring brain oxygen levels. Comparing readings with direct blood measurements from the artery (SaO2) and jugular vein (SjvO2) helps us assess performance, particularly when arterial oxyhemoglobin saturation (SaO2) is between 70% and 100%.

Drug-Induced Apnea Testing

Using medications like Propofol and Remifentanil, we induce temporary breathing pauses (apneic events) to simulate opioid-related respiratory slowdown. The aim is to evaluate how effectively the oximeter combination can identify this reduced oxygen condition, crucial for monitoring patients in these specific medical situations.

Transcutaneous Carbon Dioxide Sensor Testing

We’re testing transcutaneous carbon dioxide sensors to see how well they monitor carbon dioxide levels in the blood. We’re examining their performance alone and when used with pulse oximetry and desaturation of oxyhemoglobin (a component of the blood that carries oxygen).

Neonatal Accuracy Testing

To ensure pulse oximeters work effectively in diverse scenarios, we test them on neonates undergoing heart surgery, where physiological conditions change. Multiple blood samples are taken, providing insights into oximeter performance amidst rapidly changing variables.

Profound Hemodilution Study

This study aims to assess the accuracy of new non-invasive pulse oximeters, measuring total blood hemoglobin without skin puncture. By temporarily removing and re-infusing blood in healthy volunteers (isovolemic hemodilution), we induce changes in hemoglobin concentration.

Advanced Data Encryption Options for Secure Data Transfer

Protect sensitive study data with advanced encryption solutions designed for secure file transfer, storage, and collaboration. Multiple encryption options are available to meet sponsor requirements and organizational security standards.

Remote Video Conference Observation of Studies

Observe study procedures in real time from anywhere in the world through secure video conferencing. Sponsors, CRCs can remotely view study activities without the need for onsite travel, enabling efficient oversight while reducing logistical burdens. Please check with us to see if this service is possible for your study design.

Blood Pressure Study

Our blood pressure studies generate high-fidelity cardiovascular datasets under carefully controlled laboratory conditions in healthy human participants. The study is designed to support the development and verification of non-invasive or invasive blood pressure monitoring technologies.

During the study, participants are continuously monitored using a combination of research sensors and standard clinical monitoring equipment, including ECG, pulse oximetry, blood pressure cuffs, and an arterial pressure catheter. Assessments of cardiac output, systemic vascular resistance and other variables can also be captured. Our team of anesthesiologists and intensivists use pharmacologic and/or exercise to induce temporary, controlled increases and decreases in blood pressure, heart rate, cardiac output and systemic vascular resistance while maintaining continuous clinical oversight and predefined safety limits.

The study includes a screening evaluation followed by a 3–4 hour data collection session, which may be completed during the same visit. The resulting reference-quality datasets enable sponsors to evaluate sensor performance, develop algorithms, and validate new cardiovascular monitoring technologies with confidence.

Please reach out to us via email (hypoxialab@ucsf.edu) if you are interested in this service.

The UCSF Hypoxia Lab is one of the world’s most experienced centers for human physiologic testing. As a nonprofit, university-based laboratory, we prioritize scientific rigor, participant safety, and leverage UCSF’s unique medical and data science resources, while maintaining a strong commitment to cost-effective and efficient research practices.

Our lab brings more than 60 years of experience conducting controlled physiologic studies and has safely enrolled more than ten thousand participants. We are able to conduct not only controlled laboratory studies in healthy volunteers but also real-world studies in adult and pediatric patients at the UCSF Medical Center, enabling comparisons across the full spectrum of physiologic conditions.

The UCSF Hypoxia Lab has developed many of the protocols that have become industry standards for evaluating physiologic monitoring technologies, and we continue to innovate new methods to improve the accuracy and equity of devices. Our expanding collaboration with the Uganda Hypoxia Lab also enables faster study timelines and larger, more diverse participant populations than most testing centers can provide.

In addition to conducting regulatory-grade verification studies, we conduct regulatory science research that enables us to offer methods that we anticipate will become industry and regulatory standards. We also maintain an advanced data science and analytics team that can generate insights beyond what is required for regulatory clearance—helping teams understand device performance and improve their technologies. We create custom, secure interactive workflows and data visualization tools to make study data accessible and actionable for R&D team (beyond the reporting used for regulatory clearance). Our studies also leverage adaptable, custom-built hardware and software systems designed specifically for the unique needs of each study.

Together, these capabilities make the UCSF Hypoxia Lab a unique partner for companies seeking rigorous, innovative, and globally relevant evaluation of physiologic monitoring technologies.

The sponsor is responsible for providing their pulse oximeters and placing them on the research subjects. They also must collect their device’s data and provide that to the lab after study completion (if requesting a report for FDA). If a sponsor would like other secondary data (eg. arm circumference, a different skin pigment scale), they should collect this data themselves with permission from the participants. Sponsor staff must be present during the study.

This depends on the study sponsor’s preference. Since 2013 the FDA requires that a validation study should have participants with a range of skin pigmentations, including at least 2 darkly pigmented participants or 15% of the subject pool depending on the study size. Based on anticipated changes to FDA and ISO regulatory frameworks, we currently aim to schedule at least 24 participants per device evaluation with at least 25% of the participants in the light, medium and dark groups from the Monk Skin Tone Scale and the individual typology angle.

Up to six participants can be studied in a single day, with 20-30 arterial blood samples from each subject. This number may vary if the device under test requires additional setup time (e.g., signal stabilization and optimization) or if the sponsor requires additional metrics (e.g., surface temperature and arm circumference).

For a verification study for regulatory purposes, you should consider a minimum of 10-12 participants in order to have enough data to submit an FDA 510k report in accordance with 2013 guidance. Of note, as of 2026 both ISO and FDA are updating their guidance and standards, and a minimum of 24 participants if not more is likely to be needed.

Clinical research is the way doctors and scientists learn about new ways to prevent and treat illness in people. Clinical research is the fastest way to find answers to the many questions about health and disease. It is sometimes called “human research” or a “clinical trial.”

At UCSF most research that needs human volunteers happens in hospitals and clinics. Other types of research studies take place in a doctor’s office, a treatment center, a nursing home, or a community clinic. Some studies involve asking questions in an office, by telephone, or even on the street or in a home.

There are different ways volunteers participate in clinical research. For example:

Volunteers participate by filling out questionnaires about their health, or they may answer questions about their health in an interview with researchers.
Volunteers may be asked to donate specimens such as a tube of blood or a tissue sample for a clinical research study.
Volunteers may participate in clinical research to find out if new treatments, or new ways of using known treatments, are safe and effective to use in people.

The faculty, staff and doctors at UCSF are dedicated to research and are deeply committed to finding new ways to prevent and treat illness and improve the health of patients who come to the University Medical Center for healthcare.

A typical study session lasts around 1 to 2 hours.

30-40 minutes of this time includes blood sample data collection and breathing on the circuit, while the remainder of the time is allocated to pre- and post-study tasks (i.e., signing the consent and skin color data collection beforehand, waiting ~10 minutes after study to apply pressure to wrist and ensure subject experiences no adverse effects).

According to the UCSF Institutional Review Board (IRB) website, an IRB is “a committee — operating under federal regulations, state laws and institutional policy — that reviews research involving human subjects to ensure the ethical and equitable treatment of those subjects.”

All human subject research must be reviewed and approved by the IRB at UCSF or the Uganda Heart Institute REC in Uganda in order to be carried out. All of the Hypoxia Lab studies have received IRB or REC approval prior to being conducted.

According to the World Health Organization’s (WHO) website, clinical trials are “a type of research that studies new tests and treatments and evaluates their effects on human health outcomes.” This may include the testing of medical interventions such as “drugs, cells and other biological products, surgical procedures, radiological procedures, devices, behavioral treatments and preventive care.

Clinical trials are carefully designed, reviewed and completed, and need to be approved before they can start. People of all ages can take part in clinical trials, including children.”