Introduction

The annual Consumer Electronics Show (CES), held in Las Vegas, NV, USA, is the world’s most extensive and influential electronics exhibition, serving as a global stage for showcasing cutting-edge technologies and innovative consumer electronics [1]. Companies worldwide have attended CES to unveil groundbreaking products and services, introducing revolutionary technologies encompassing artificial intelligence (AI), autonomous driving, virtual reality, and smart home technology. Exhibitions garner extensive attention from technology experts, the media, and consumers worldwide, thereby solidifying the role of these exhibits as vital platforms for shaping future trends in electronic products. The inaugural CES occurred in New York City in 1967. Since 1995, it has been held annually in Las Vegas, with CES 2025 occurring from January 7 to January 10. In our capacity as observers of CES 2025, we present a summary report highlighting several emerging technologies with significant potential in the medical field. Among these, AI and sensor technologies are particularly prominent and emerging as the most widely applied technologies in the healthcare sector.

Artificial intelligence

AI has achieved a high level of recognition as a technology with extensive applications in healthcare, and its utility is now widely accepted [2,3]. However, practical applications of AI remain largely in the research phase, and the commercialization of AI-based programs and products remains restricted for general use. Several products that have recently been launched or are nearing release were showcased at CES 2025.

For example, a product employing deep learning algorithms trained on biometric data collected from individuals to recommend the most suitable binaural beat stimulation for each user was presented. This product was designed to match the shape of wireless earphones and employs embedded photoplethysmography (PPG) sensors to measure the pulse rate and body temperature of users [4,5]. The functionality of this device extends beyond mere recommendations by predicting additional data points, including electroencephalogram results and pulse pressure. Overall, this device enabled the delivery of binaural beat stimulation at an optimal frequency to promote mental relaxation and enhance sleep quality.

Another exhibit featured a program that utilized an AI algorithm based on natural language processing. This program was designed to identify and examine the speech patterns of users to evaluate language impairment and subsequently devise bespoke treatment plans. A similar device employing AI to analyze snoring sounds was also introduced. This product uses an airbag within a pillow to adjust the user’s head position, thereby opening the airway to reduce or eliminate snoring.

Another showcased product employed AI to predict blood pressure by analyzing pulse rate, body temperature, and oxygen saturation (SpO₂) measurements obtained during sleep using a remote PPG (rPPG) sensor affixed above the bed. The rPPG sensor is a contactless method for measuring heart rate and other vital signs by analyzing subtle color changes in the skin captured using a camera. This allows physiological monitoring without physical contact with the subject [5,6].

However, it should be noted that a direct comparison of the data obtained from the rPPG sensor with the ground truth data could not be reliably established. Furthermore, an AI-powered device that uses photographs to calculate food calories was developed. However, its functionality is limited to the analysis of children’s meals (Fig. 1), likely because of insufficient training data to account for a greater variety of adult foods.

Despite the advent of many AI-based products, these devices exhibit deficiencies in the degree of accuracy and substantiation required for their application in clinical settings. Consequently, the devices exhibited at the CES are better suited for community-level applications, such as residential settings and screening purposes, as opposed to medical applications in hospitals.

Sensors

Sensors can be classified into two categories: contact and noncontact [7]. Although contact-type sensors generally exhibit higher accuracy under static conditions than their noncontact counterparts, the necessity of maintaining physical contact during data collection can render them less convenient for certain applications. In contrast, noncontact sensors can assess or measure physical properties or conditions without requiring direct physical contact [8].

A predominant tendency among the contact-type sensors presented at CES 2025 was to curtail physical contact to mitigate user discomfort or ensure the unobtrusive maintenance of contact conditions. For example, one product featured small sensors placed on both thumbs to record electrocardiograms (EKGs). Another product was a ring-shaped sensor that, when worn on a finger, utilizes an embedded PPG to measure parameters such as pulse rate, SpO₂, pulse pressure, blood glucose levels, and body temperature. Furthermore, a product that enables effortless blood pressure monitoring through a sensor affixed to a user’s watch was introduced. This device employed a combination of piezoelectric sensor and near-infrared spectroscopy to measure blood pulses [9]. However, the mechanisms underlying these sensors remain unclear for obtaining reliable and stable results. Conventional PPG sensor platforms such as rings and watches encounter challenges in completely blocking ambient light. Although many companies which exhibited at CES 2025 have demonstrated the potential of these technologies, further elucidation is required to fully understand their operational mechanisms.

The development of noncontact sensors has been predominantly driven by radar and rPPG technologies. For example, a radar sensor designed to detect falls was presented. Another radar-based sensor developed to measure respiration and heart rates by detecting the movements of internal organs was also showcased (Fig. 2). Radar-based vital sign detection is a technique that utilizes radio waves to measure physiological signals such as respiration and heartbeat. This process involves detecting body movements to generate organ movements using radio waves, thereby enabling contactless monitoring even through clothing or in low-light environments [10]. However, these sensors are unable to measure these parameters while the user is in motion. Thus, the utilization of these sensors is constrained to circumstances in which the subject remains stationary, such as during sleep or in motionless or bedridden patients. A radar sensor designed to monitor driver respiration and heart rate inside a vehicle was also presented. This device uses rPPG technology to measure pulse rate by detecting subtle facial color changes.

Despite noteworthy advancements in the field, the sensors for healthcare applications exhibited at CES 2025 predominantly have limited accuracy, rendering them unsuitable for clinical use. Another salient concern is the absence of field demonstrations to substantiate the reliability of these products through an accurate comparison with ground truth. This limitation is likely to constrain their use in community settings for screening. Technological advancements are imperative to enable accurate measurements, even during motion, to expand the scope of noncontact sensor applications. Despite the dearth of published results, none of the sensors at CES 2025 are currently available as marketable products. Most sensors claim to utilize deep learning to manage training data to enhance accuracy; however, the technical specifics and validity of these claims were not provided.

Fusion of artificial intelligence and sensors

Some devices showcased at CES 2025 integrated AI and sensor technologies to achieve functionality. For example, a novel electronic stethoscope equipped with a microelectromechanical system has been developed to detect internal body sounds. This device, which is ergonomically designed to be held in the hand and positioned on the chest, facilitates monitoring of cardiac and respiratory sounds. The company also developed an accompanying deep learning model that allows users to perceive heart and breathing sounds as if listening through a real stethoscope. This was achieved by training the model to differentiate the sensed heart and breath sounds and to output authentic stethoscope sounds and EKG findings by implementing deep learning methodologies. The compact design of this device, which facilitates portability and ease of use, is ensured by its handheld form. This device also uses a smartphone application to visually present EKG findings, heart rates, and breathing data. Finally, the system incorporates functionality that facilitates the identification of cardiac abnormalities through the analysis of heartbeats and respiratory sounds.

Other products

Among products that utilize technologies other than AI and sensors, a particularly impressive device was a medical tool that utilizes a laser instead of a disposable needle to painlessly draw blood while measuring blood glucose levels. A salient feature of this innovation is the ability to transmit glucose readings to designated applications, thereby facilitating real-time data monitoring and management. Another notable innovation was a belt-type wearable robot weighing just 1.6 kg, which is remarkably lightweight for robotic devices. This device assists wearers in ambulating by elevating their legs, thereby enhancing comfort, and it further supports gait training by applying resistance during movement. Ease of use, convenience, and affordability were identified as the key benefits of this device.

Conclusion

Overall, the spectrum of devices presented at CES 2025 demonstrates the innovative potential of electronic technology for advancing healthcare and improving its quality. The most notable innovations showcased at the event pertained to AI and sensor technologies. Specifically, AI-powered products have demonstrated potential for personalized health management and treatment, whereas sensors utilizing radar and rPPG technologies can enable noncontact measurement of vital signs, such as respiratory and pulse rates. Despite the current limitations in the accuracy of these technologies, which make them unsuitable for clinical applications, they hold considerable promise as screening tools in home and community settings. However, to ensure widespread adoption, it is essential that measurement techniques such as those for blood pressure and blood glucose are supported by extensive experimental evidence. Furthermore, the integration of advanced technologies, such as laser-based blood collection systems and wearable robots designed to assist with gait, is anticipated to enhance convenience and improve patient quality of life.

Article information

Conflicts of interest

Min Cheol Chang has been a Deputy Editor of Journal of Yeungnam Medical Science since 2025. He was not involved in the review process of this manuscript. There are no other conflicts of interest to declare.

Funding

This study was supported by a National Research Foundation of Korea Grant funded by the Korean government (No. 00219725).

Author contributions

Conceptualization, Data curation, Investigation, Methodology, Formal analysis, Funding acquisition: JRY, MCC; Supervision: MCC; Writing-original draft: JRY, MCC; Writing-review & editing: JRY, MCC.

Fig. 1.

Photograph of an artificial intelligence-powered device designed to estimate calories in food.

Fig. 2.

Photograph of radar sensors designed to detect falls and vital signs.

References

• 1. Marrouche NF. New consumer guidance on wearable devices from the Consumer Electronics Show 2020. Heart Int 2020;14:9–10.ArticlePubMedPMC
• 2. Chang MC. Development of an automated foot contact area measurement program for podoscopes using ChatGPT-4: a case report. J Yeungnam Med Sci 2025;42:13.ArticlePubMedPMCPDF
• 3. Kong HJ. Classification of dental implant systems using cloud-based deep learning algorithm: an experimental study. J Yeungnam Med Sci 2023;40(Suppl):S29–36.ArticlePubMedPMCPDF
• 4. Khan M, Pretty CG, Amies AC, Elliott R, Shaw GM, Chase JG. Investigating the effects of temperature on photoplethysmography. IFAC Pap OnLine 2015;48:360–5.Article
• 5. Xiao H, Liu T, Sun Y, Li Y, Zhao S, Avolio A. Remote photoplethysmography for heart rate measurement: a review. Biomed Signal Process Control 2024;88:105608.Article
• 6. Scardulla F, Cosoli G, Spinsante S, Poli A, Iadarola G, Pernice R, et al. Photoplethysmograhic sensors, potential and limitations: is it time for regulation?: a comprehensive review. Measurement 2023;218:113150.Article
• 7. Naresh V, Lee N. A review on biosensors and recent development of nanostructured materials-enabled biosensors. Sensors (Basel) 2021;21:1109.ArticlePubMedPMC
• 8. Choo YJ, Moon JS, Lee GW, Park WT, Chang MC. The applicability of noncontact sensors in the field of rehabilitation medicine. J Yeungnam Med Sci 2024;41:53–5.ArticlePubMedPMCPDF
• 9. Guo CY, Chang HC, Wang KJ, Hsieh TL. An arterial compliance sensor for cuffless blood pressure estimation based on piezoelectric and optical signals. Micromachines (Basel) 2022;13:1327.ArticlePubMedPMC
• 10. Park JE, Lee GH, Lee IS, Yang JR. Heart rate extraction technique with mitigation of respiration harmonic for bio-radar sensors. IEEE Sens J 2025;25:929–39.Article

Figure & Data

References

Citations

Citations to this article as recorded by