Oral monitoring plays an essential role in preventing and diagnosing systemic diseases through saliva in the mouth. Dietary monitoring is also crucial to reduce the likelihood of chronic diseases such...

Dietary monitoring is vital in healthcare because knowing food mass and intake (FMI) plays an essential role in revitalizing a person’s health and physical condition. In this study, we report the development of a highly sensitive ring-type biosensor for the detection of FMI for dietary monitoring. To identify lightweight food on a spoon, we enhance the sensing system’s sensitivity with three components: (1) a first-class lever mechanism, (2) a dual pad sensor, and (3) a force focusing structure using a ring surface having protrusions. As a result, we confirmed that, as the food arm’s length increases, the force detected at the sensor is amplified by the first-class lever mechanism. Moreover, we obtained 1.88 and 1.71 times amplification using the dual pad sensor and the force focusing structure, respectively. Furthermore, the ring-type biosensor showed significant potential as a diagnostic indicator because the ring sensor signal was linearly proportional to the food mass delivered in a spoon, with R2 = 0.988, and an average F1 score of 0.973. Therefore, we believe that this approach is potentially beneficial for developing a dietary monitoring platform to support the prevention of obesity, which causes several adult diseases, and to keep the FMI data collection process automated in a quantitative, network-controlled manner.

Body shape and curvature are vital criteria for judging health. However, few studies exist on the curvature of the body. We present a skin-interactive electronic sticker that digitally decodes the epidermis deformation in a hybrid cartridge format (disposable bandages and non-disposable kits). The device consists of two functional modes: (1) as a thin electronic sticker of 76 μm thickness and a node pitch of 7.45 mm for the measurement of body curvature in static mode, and (2) as a wrist bandage for the deciphering of skin wave fluctuations into a colored core-line map in dynamic mode. This method has high detection sensitivity in the static mode and high accuracy of 0.986 in the dynamic mode, resulting in an F1 score of 0.966 in testing by feedforward deep learning. The results show that the device can decipher 32 delicate finger folding gestures by measuring skin depths and positions via image segmentation, leading to an optimal core line in a color map. This approach can help provide a better understanding of skin wave deflection and fluctuations for potential wearable applications, such as in delicate skin-related gesture control in the metaverse, rehabilitation programs for the brain-degenerate, and as a detector of biophysical state relating to body shape and curvature in the field of digital medicine.