Measuring vital signs can reveal the state of body functioning and help to detect a health problem. In the state-of-the-art, numerous methods and devices are available for measuring vital signs. However, with the advent of artificial intelligence, new methods have been proposed that employ this technology. This paper aims to highlight the recent methods and devices based on artificial intelligence and novel techniques for measuring vital signs and processing algorithms. We analyzed 122 papers and classified them into six categories: (i) body temperature, (ii) blood oxygen saturation, (iii) heart rate monitoring, (iv) respiratory rate, (v) blood pressure, and (vi) simultaneous vital sign measurements. The Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) methodology was used for the search and selection of scientific papers. The criteria to guide the scope of the review were defined with the Population, Intervention, Comparison, Outcomes, and Context (PICOC) methodology. The review highlighted significant efforts to develop and implement contactless, non-invasive devices for continuous monitoring outside clinical environments. It also revealed clear pathways for integrating AI at different stages of measurement and signal processing methods.

Because of the increased prevalence of chronic diseases, older adults frequently take many medications. However, adhering to a medication treatment tends to be difficult. The lack of medication adherence can cause health problems or even patient death. This paper describes the methodology used in developing a mobile augmented reality (MAR) pill box. The proposal supports patients in adhering to their medication treatment. First, we explain the design and construction of the automatic pill box, which includes alarms and uses QR codes recognized by the MAR system to provide medication information. Then, we explain the development of the MAR system. We conducted a preliminary survey with 30 participants to assess the feasibility of the MAR app. One hundred older adults participated in the survey. After one week of using the proposal, each patient answered a survey regarding the proposal functionality. The results revealed that 88% of the participants strongly agree and 11% agree that the app is a support in adhering to medical treatment. Finally, we conducted a study to compare the time elapsed between the scheduled time for taking the medication and the time it was actually consumed. The results from 189 records showed that using the proposal, 63.5% of the patients take medication with a maximum delay of 4.5 min. The results also showed that the alarm always sounded at the scheduled time and that the QR code displayed always corresponded to the medication that had to be consumed.

Segmenting brain tumor subregions in multimodal MRI is difficult due to severe class imbalance and scarce annotated data, and current state-of-the-art models require 21.3–31 million parameters to reach a whole-tumor Dice of 0.906–0.921. We propose a parameter-efficient 2D encoder–decoder coupling a Fourier Neural Operator (FNO) and a Vision Transformer (ViT), trained with BraTSPipeline, which raises throughput from 257 to 16,040 slices per epoch, and two composite loss functions penalizing false negatives 2.3× more than false positives. On BraTS 2020, the 2D variant (5.6 M parameters) achieves whole-tumor (WT), tumor-core (TC), and enhancing-tumor (ET) Dice of 0.8985, 0.8263, and 0.7568 on the 56-case held-out test set, using 3.8–5.6× fewer parameters than published architectures. Encoding three consecutive axial slices as 12 channels (2.5D, 20.5 M parameters) raises WT to 0.9117, within 0.010 of the best published 2D result on BraTS 2020, Mod-R2AU-Net (WT = 0.921); a 4.6 M-parameter ablation without the ViT reaches WT of 0.9106 and TC of 0.8428, while the ViT adds 0.043 ET Dice.

Light Detection and Ranging (LIDAR) is a remote sensing technique that measures properties of backscattered light in order to obtain information of a distant target. This work presents a method to improve the signal-to-noise ratio by 8 dB with respect to the direct detection of the backscattered signal of a LIDAR system. This method consists of the measurement of the Fourier coefficients of the LIDAR signal, which is possible thanks to the novel coupling of a sequential equivalent time base sampling (SETS) circuit and a conventional lock-in amplifier that allows to measure the Fourier coefficients of the LIDAR signal, the results are comparable to noise elimination using Empirical Mode Decomposition. The feasibility of the proposal is demonstrated experimentally with mist. The method can be used to different applications of elastic-scattering LIDAR under the conditions of the devices for applied the phase sensitive detection.