Digital blood pressure monitors using the oscillometric method are less accurate than commonly assumed. These $30-50 devices inflate an arm cuff and detect pressure oscillations during deflation, using fixed percentage thresholds (e.g., 40% and 80% of oscillation amplitude) to estimate systolic and diastolic values — an approximation rather than a direct measurement. Category A monitors are only required to land within ±15 mmHg for 85% of readings. To investigate this, Milos Rasic built the Open Cardiography Signal Measuring Device, an open-source platform that combines arm cuff pressure sensing with PPG, ECG, and digital stethoscope inputs to test alternative measurement algorithms and evaluate off-the-shelf monitor accuracy.
Nguồn: https://hackaday.com/2026/07/06/hackaday-europe-2026-is-your-blood-pressure-monitor-lying-to-you. 8sync News chỉ tóm tắt và dẫn link; bản quyền nội dung thuộc tác giả và nguồn gốc.
MIT researchers have developed a portable ultrasound system aimed at making breast cancer screening more accessible and frequent. The new system features a backing layer added to the ultrasound transducer that improves image resolution and reduces noise, plus an adaptive beamforming algorithm that compensates for varying sound speeds across tissue types, yielding up to 10% resolution improvement. A computer-vision-based user interface guides non-expert users to position the probe correctly, enabling consistent longitudinal monitoring. In trials, untrained volunteers successfully located embedded targets at higher rates than with traditional probes. The team plans to develop a smartphone-compatible version and is exploring commercialization, with potential applications beyond breast cancer to ovarian cancer, endometriosis, and fetal monitoring.
Consumer wearables report different heart rates for the same person due to several compounding factors. All wrist and finger wearables use photoplethysmography (PPG), but sensor placement matters — finger-based devices like Oura Ring sit closer to surface arteries and move less during sleep, giving them an accuracy edge over wrist devices. Sampling rate also plays a major role: WHOOP samples 26 times per second continuously, while Apple Watch and Garmin use periodic or adaptive sampling. Beyond hardware, proprietary algorithms translate raw light signals into heart rate values, and software updates alone can shift reported numbers without any hardware change. Additional variables include skin tone (melanin absorbs more light, reducing signal quality), tattoos over sensor areas, and device fit on the wrist. During exercise, motion artifacts further diverge readings between brands. The practical takeaway is that cross-device comparisons are unreliable; what matters is consistent trends on a single device over time.