Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Non-contact pulse oximetry for MRI
Pulse oximetry during Magnetic Resonance Imaging (MRI) helps to monitor the well-being of the patients and to obtain information about their physiology. However, the standard oximetry finger clip complicates the process and therefore it is the aim of the project to build a contactless pulse oximeter that can be integrated into a novel low-field MRI scanner. To this end, an MRI compatible camera will be deployed to illuminate the face and to quantify the reflected light intensity on volunteers. Machine learning algorithms will be used to derive physiology triggers to e.g. control the MRI data acquisition process.
Keywords: Magnetic resonance imaging, contactless oximetry, signal processing, machine learning
**Background**
Pulse oximeters are devices of high clinical importance because they allow to measure the arterial oxygen saturation and thus, a crucial vital sign. A lack of oxygen, which could be fatal within minutes, can be detected by these instruments. During a critical situation, they will sound an alarm, which enables the staff to resolve the problem. Currently, the sensor of pulse oximeters is a clip, which is attached to a finger. It measures the arterial oxygen saturation by shining red and near-infrared light through the finger. The arterial blood concentration changes with systole and diastole. The spectrum of the arterial blood can be obtained simply be subtracting the spectrum during systole from the spectrum during diastole. Pulse oximeters are ubiquitous in hospitals and in ambulances.
They are also important during Magnetic Resonance Imaging (MRI) to monitor the well-being of the patients and to obtain information about their physiology. However, the finger clip complicates the process and therefore we aim to build a contactless pulse oximeter that is part of the scanner. This can be achieved by employing an MRI compatible camera and an illumination with appropriate wavelengths. Thus, the light will illuminate e.g. the face and the camera will quantify the reflected light intensity.
**Tasks of the project**
1. Perform a literature search to summarize previous work in this field to define the state of the art.
2. Develop and discuss different concepts for the new contactless pulse oximeter.
3. Compile a specification of the requirements to realize the concept.
4. Implement the concept using prototype hard- and software.
5. Test the device and compare it to a standard clinical pulse oximeter in a few subjects.
6. Compile a report and documentation to enable future students and staff to continue the work.
**What is needed?**
Electronic engineering/programming knowledge.
**Background** Pulse oximeters are devices of high clinical importance because they allow to measure the arterial oxygen saturation and thus, a crucial vital sign. A lack of oxygen, which could be fatal within minutes, can be detected by these instruments. During a critical situation, they will sound an alarm, which enables the staff to resolve the problem. Currently, the sensor of pulse oximeters is a clip, which is attached to a finger. It measures the arterial oxygen saturation by shining red and near-infrared light through the finger. The arterial blood concentration changes with systole and diastole. The spectrum of the arterial blood can be obtained simply be subtracting the spectrum during systole from the spectrum during diastole. Pulse oximeters are ubiquitous in hospitals and in ambulances. They are also important during Magnetic Resonance Imaging (MRI) to monitor the well-being of the patients and to obtain information about their physiology. However, the finger clip complicates the process and therefore we aim to build a contactless pulse oximeter that is part of the scanner. This can be achieved by employing an MRI compatible camera and an illumination with appropriate wavelengths. Thus, the light will illuminate e.g. the face and the camera will quantify the reflected light intensity.
**Tasks of the project** 1. Perform a literature search to summarize previous work in this field to define the state of the art. 2. Develop and discuss different concepts for the new contactless pulse oximeter. 3. Compile a specification of the requirements to realize the concept. 4. Implement the concept using prototype hard- and software. 5. Test the device and compare it to a standard clinical pulse oximeter in a few subjects. 6. Compile a report and documentation to enable future students and staff to continue the work.
**What is needed?** Electronic engineering/programming knowledge.
Not specified
Prof. Martin Wolf
Biomedical Optics Research Laboratory (BORL)
www.usz.ch/fachbereich/neonatologie/forschung/borl
martin.wolf@usz.ch
Prof. Sebastian Kozerke
Institute for Biomedical Engineering (IBT)
www.cmr.ethz.ch
kozerke@biomed.ee.ethz.ch
Prof. Martin Wolf Biomedical Optics Research Laboratory (BORL) www.usz.ch/fachbereich/neonatologie/forschung/borl martin.wolf@usz.ch
Prof. Sebastian Kozerke Institute for Biomedical Engineering (IBT) www.cmr.ethz.ch kozerke@biomed.ee.ethz.ch