What do biomedical engineering majors study? Picture a wrist monitor that reads 78 beats per minute while you sit, then struggles to give a steady reading as you swing your arms. To understand why, you need to know what makes a pulse, how a sensor detects it, and how software decides which changes to count. To make the device useful, you also need to decide what it should show when it cannot get a good reading. Those are problems from physiology, electronics, statistics, and design.
Physiology explains what the device is trying to measure. When the heart pumps blood into the arteries, a pressure wave travels through them. At the wrist, changes in local blood volume over the pulse cycle change how much light returns to a sensor. This technique is called photoplethysmography, or PPG. Many wrist monitors use it to estimate pulse rate.
A pulse signal is not the same as an electrocardiogram, or ECG. An ECG records electrical activity associated with the heart. An optical wrist sensor measures changes in reflected light associated with the pulse. Those signals can both help estimate a rate, but they describe different events. A physiology class helps you explain why a sensor can pick up a pulse at the wrist even though the heartbeat starts in the chest.
Heart monitoring is only one example. Depending on the program, students may also study how muscles generate force or how blood flows through vessels. A course list such as the Georgia Tech biomedical engineering degree requirements shows how one university combines engineering, math, and life science. When comparing colleges, look at which courses are required and which are electives rather than assuming every program covers the same devices.
In an optical wrist monitor, a light-emitting diode shines light into the skin, and a photodiode converts some of the returning light into an electrical current. Circuitry turns that small current into a voltage the device can measure. The device then converts its measurements into a series of numbers that software can process. In an electronics lab, you might use a circuit and a light sensor to see how changing the light changes the measured voltage.
The hard part is that a heartbeat is not the only thing that changes the light reaching the sensor. If the watch shifts when you swing your arm, the contact between the sensor and skin changes. That can distort the signal or produce patterns the software mistakes for pulses. Filtering can reduce some unwanted changes, but it is harder to separate movement from a pulse when both produce changes at similar rates. A better strap fit or a change in sensor placement may help more than another software adjustment.
Suppose you have a program that marks peaks in the optical signal and uses their spacing to estimate a rate. A statistics assignment could ask you to compare its readings with a reference, such as a rate calculated from an ECG recorded at the same time. You could compare matched intervals at rest and during movement. Matching the intervals matters because the heart rate itself can change between two readings.
An average error can hide a problem. Imagine a monitor that reads 10 beats per minute too high in one interval and 10 too low in another. Its average signed error is zero, but neither estimate is right. The average absolute error, which ignores the direction of each mistake, is 10 beats per minute in this example. You would also count the intervals when the monitor gives no reading. Otherwise, a device that goes blank during difficult measurements could look better than it really is.
In a design course, you might start by defining who will use the monitor and what they need from it. A display meant to be read during exercise has different requirements from a recorder meant to store data for later review. If the optical signal becomes unreliable, the display could hold the last estimate or show that it cannot get a reading. Holding the old number keeps the display steady, but someone might mistake it for a current measurement. A prototype lets you check whether users understand the difference.
Medical training focuses on caring for patients. A physician evaluating someone who reports a racing heart may ask when it happens, review medications, examine the patient, and decide whether tests are needed. A biomedical engineer working on a monitor may instead investigate why it missed some pulses or how to show uncertainty without confusing the wearer. Both roles draw on physiology and evidence, but a biomedical engineering degree does not qualify someone to practice medicine.
You can try the signal-processing side of biomedical engineering without buying a sensor or collecting anyone's health data. Build a counter in a spreadsheet, then see how easily it confuses simulated movement with a pulse.
For one more test, change the samples on either side of a real peak to zero. Does your second rule now miss it? Write down what each rule gets wrong and describe a case where you would rather have the display report no reading than report a number. For practice making sense of biological data beyond this toy signal, explore the IMedC training and practice problems or browse the IMedC learning resources.
Do biomedical engineering majors have to take electronics?
Requirements vary by college. Some programs require circuits or instrumentation, while others offer more electronics through electives or a concentration.
Can I study biomedical engineering and apply to medical school?
Yes. Check the admissions requirements of the medical schools you are considering, since the engineering degree may not include all their prerequisite courses.
Do I need a smartwatch for the school project?
No. A spreadsheet and made-up data let you test whether a counting rule can distinguish regular peaks from an extra spike.
About the International Medicine Competition: an online medicine competition for high school students from around the world. Qualification Round 2027: submissions are open until Sunday, 7 February 2027. How the rounds work