Search tips
Search criteria 


Logo of appciLink to Publisher's site
Appl Clin Inform. 2016; 7(3): 850–869.
Published online 2016 September 14. doi:  10.4338/ACI-2016-03-RA-0042
PMCID: PMC5052554

Healthcare Applications of Smart Watches

A Systematic Review



The aim of this systematic review is to synthesize research studies involving the use of smart watch devices for healthcare.

Materials and Methods

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) was chosen as the systematic review methodology. We searched PubMed, CINAHL Plus, EMBASE, ACM, and IEEE Xplore. In order to include ongoing clinical trials, we also searched Two investigators evaluated the retrieved articles for inclusion. Discrepancies between investigators regarding article inclusion and extracted data were resolved through team discussion.


356 articles were screened and 24 were selected for review. The most common publication venue was in conference proceedings (13, 54%). The majority of studies were published or presented in 2015 (19, 79%). We identified two registered clinical trials underway. A large proportion of the identified studies focused on applications involving health monitoring for the elderly (6, 25%). Five studies focused on patients with Parkinson’s disease and one on cardiac arrest. There were no studies which reported use of usability testing before implementation.


Most of the reviewed studies focused on the chronically ill elderly. There was a lack of detailed description of user-centered design or usability testing before implementation. Based on our review, the most commonly used platform in healthcare research was that of the Android Wear. The clinical application of smart watches as assistive devices deserves further attention.


Smart watches are unobtrusive and easy to wear. While smart watch technology supplied with biosensors has potential to be useful in a variety of healthcare applications, rigorous research with their use in clinical settings is needed.

Keywords: Other clinical informatics applications, Interfaces and usability, Healthcare, Wearable device, Smart watch

1. Introduction

1.1 Background

There is little doubt that wearable technologies are entering our lives, especially amongst early adopters. Numerous technology companies have invested in developing novel wearable solutions to gain successful access into consumer markets. It was estimated that only 1% to 2% of individuals in the United States have used a wearable device, but the market is forecasted to be worth $25 billion by 2019 with smart watches taking 60% of market value [1, 2].

A wearable device can be defined as a mobile electronic device worn as an accessory or unobtrusively embedded in the user’s clothing [3]. Generally, wearable devices adopt the technologies of sophisticated biosensors and wireless data communication that allow the wearer to access and transmit information in all sectors of human endeavor. Given the functionality of miniaturized biosensors capable of wireless communication, these devices are developed to be innovative, non-invasive monitoring technologies for continuous and autonomous transmission of physiological data [4]. As these wearable devices proliferate in the clinical domain, they have the potential to provide caregivers with the information they need to improve the quality of health care, change and facilitate clinical workflow, manage and treat patients remotely, collect greater health data, and deliver more meaningful healthcare to patients [5].

For practical use, Zhang’s research group noted several key factors that should be developed in order to implement wearable devices, including miniaturization, integration, networking, digitalization, and standardization [6]. To be comfortably worn on the body, miniaturization and unobtrusiveness are considered the most important factors that can increase compliance for long-term and continuous monitoring [7]. A recent advent to the fast-growing market of wearable devices is the smart watch. With its miniaturized form factor design and computing technology, a smart watch can be worn continuously without interrupting the user’s daily activity. Although smart phones have become a part of our daily lives and might be considered to be wearable, these devices most often resides in a pocket or purse. Unlike smart phones, smart watches can be truly wearable without interrupting our daily lives, and can also serve as a readily accessible extension of the smart phone [8]. Because of the proximity to the skin, the smart watch can also be a source of physiological data derived directly from the wearer‘s body [9]. With the potential for widespread adoption in the healthcare sector, smart watches equipped with biosensors have the potential to provide important healthcare information to patients and their providers.

1.2 Significance

While there is potential for smart watch technology to gather and display important health data, to our knowledge there has been no systematic review regarding its healthcare application either in the research environment or in clinical practice.

1.3 Objectives

In this article, we aim to review the published literature regarding healthcare applications of smart watches and the ongoing research projects that have been registered in the government clinical trials website. We also discuss the potential uses and limitations of smart watches in healthcare settings.

2. Material and Methods

2.1 Literature Search

We chose the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) as the systematic review methodology [10]. A total of five databases were searched, including PubMed, CINAHL Plus, EMBASE, ACM and IEEE Xplore Digital Library. All databases were searched by using keywords “Smart Watch” or “Smartwatch”, along with the brand names of the most commonly available commercial smart watches. Additionally, searches were conducted on to include ongoing registered clinical trials. Although this review focused on healthcare applications, no reference to healthcare or application was included in the search terms to ensure a broad sweep of articles for consideration. The search terms used in PubMed were as follows and were modified to fit specific requirements of each of the databases searched.

(“smart watch“[All Fields] OR smartwatch[All Fields]) OR (“Android“[All Fields] AND “Wear“[All Fields]) OR (“Apple“[All Fields] AND “Watch“[All Fields]) OR (“Moto“[All Fields] AND “360“[All Fields]) OR (“Samsung“[All Fields] AND “Gear“[All Fields]) OR (“Pebble“[All Fields] AND “Watch“[All Fields]) OR (“Garmin“[All Fields] NOT (“GPS“[All Fields] OR “Global Positioning System”[All Fields])) NOT (“Comment“[Publication Type] OR “Editorial“[Publication Type] OR “Review”[Publication Type])

We ran our search on December 1, 2015. We did not limit the year in the search terms, since the smart watch and its applications in the healthcare domain are relatively new. Additionally, we conducted a manual review of the citations included in the articles retrieved.

2.2 Article Selection

One of the authors conducted an initial screen on the retrieved records. Duplicated articles were eliminated and additional records were excluded after reviewing individual titles and abstracts. A second author then reviewed the included studies. The retrieved full-text articles were evaluated for eligibility by two independent investigators. Reviewers were blinded to each other’s assessments. Discrepancies about article inclusion were then resolved through discussion with other team members. After excluding irrelevant studies, the remaining studies were selected for final review. To be included in the final review, studies had to be:

  1. published in peer-reviewed journals either as original articles or as conference proceedings, or be registered as an ongoing study in the official Clinical Trials website maintained by the National Library of Medicine (NLM) (i.e.,
  2. featuring smart watch or smartwatch as the primary subject of study or a main component of the study methodology.
  3. targeted toward the clinical application of specific diseases of interest or individuals with specific healthcare demands.
  4. written in English.

We excluded those articles that were not considered original research, such as letters to the editor, comments, or reviews. Because this review focused on smart watches, wearable wrist devices without the functionality of watches were also excluded. We also excluded smart band devices that solely tracked activity or fitness.

2.3 Data Extraction

After the articles were selected for final review, they were randomly assigned to two investigators who extracted data and entered into a free online spreadsheet (Google Sheets). Data extracted included: authors, year of publication, publication type, study design, target population, number of participants, study aims, study intervention, technology-related findings, platform and/or type of smart watch, type of sensors used, and article title. We also extracted information from each article according to whether the study described the use of human-computer interaction, user-centered design, or pre-implementation usability testing as part of their main study interventions or findings. Finally, discrepancies about the contents of the extracted data were resolved through team discussion.

3. Results

Initially, 356 studies were identified through database searching. After excluding duplicated records, 325 records were eligible for screening. There were 292 records that did not meet our inclusion criteria based on the screen. A total of 33 studies were included to be evaluated for eligibility. Full text records were retrieved and reviewed by two independent assessors. After excluding irrelevant studies, 24 articles were selected for final review, including 7 original studies, 2 conference papers, 13 conference proceedings, and 2 ongoing clinical trials. The study selection process is depicted in ► Figure 1. The complete description of the included studies is shown in ► Table 1.

Fig. 1
The study selection process.
Table 1
Description of All Articles about Healthcare Related Smart Watch Research

Of the 24 records selected, the most common year published, presented, or registered was 2015 (19, 79%), followed by 2014 (4, 17%). There was only one article published earlier, in late 2013 (4%). In terms of the publication type, 13 (54%) were published as conference proceedings and seven (29%) as journal articles. With respect to study design, the largest number of studies (13, 54%) utilized experimental designs in which machine learning was used to create annotated datasets for classification or pattern recognition to model a smart watch intervention for a target population, followed by experimental designs with control groups (5, 20%) to investigate the effect of the smart watch intervention on specific outcomes. There were no clinical trials published. However, in we identified two studies underway involving smart watches (2, 8%). For studies that have been completed and published, the number of participants or patients ranged from 1 to 143. The highest number of studies were conducted in the USA (10, 42%), followed by three studies in Germany (13%) and two studies in the United Kingdom (8%). The remaining nine studies were conducted in different countries around the world.

With respect to the target population, six studies (25%) focused on smart watch use among the elderly, either for health monitoring or in a smart home environment, and five studies (21%) focused on patients with Parkinson’s disease (PD). The third and fourth largest groups of studies focused on food and diet monitoring (4, 17%) and on medication adherence monitoring in patients with chronic diseases (3, 13%). Although there were dozens of smart watches to choose from, the most commonly used platforms for healthcare research were those involving the Android Wear (11, 46%). Among those, the most commonly used brand was the Samsung Galaxy Gear (6, 25%) followed by the Pebble Smartwatch (4, 17%). Although most studies featured the smart watch as the primary subject of study, seven studies (29%) utilized both a smart watch and a smart phone as main components of the study methodology. Study characteristics, including study design, target population, and platform used, etc., are summarized in ► Table 2. Number of publications and types of study design in terms of the target population is shown in ► Figure 2 (A). For the most commonly used study methodology, the experimental study of machine learning, the number of publications with respect to different target population is presented in ► Figure 2 (B).

Fig. 2
(A).Number of publications (Y-axis) and types of study design in terms of the target population (X-axis). (B).Number of publications (Y-Axis) with respect to different target population (X-axis) for study design using experimental study of machine learning. ...
Table 2
Characteristic of Selected Articles

In terms of utilizing the accelerometer or gyroscope functionalities that smart watches general exhibit, most of the selected studies used at least one of these functionalities as the main concept of applications for their studies (16, 67%). Of them, five studies (21%) used the combination of an accelerometer and a gyroscope [11, 12, 28–30]. Seven studies did not utilize any sensor in their study intervention [14–15, 18, 23–24, 27, 33]. Instead, smart watches were used as assistive devices for patients with specific needs via their screen or voice as input or reminders. One study utilized physiological sensors to monitor activity in the elderly by recording heart rate and skin temperature [31].

In most of the studies (18, 75%) there was no mention of human-computer interaction, user-centered design, or pre-implementation usability testing as part of their study design or intervention. However, two studies utilized user-centered design during the design phase [15, 22], one study had a brief evaluation of the user interface [27], and three studies mentioned usability testing in the context of future work [12, 18, 20].

4. Discussion

Our review of the literature revealed that, since late 2013, there were 24 studies involving smart watches in healthcare applications that met our inclusion criteria. Given their recent appearance on the commercial market, it is not surprising that the majority of these studies were published in 2015. This review discloses a wide variation in study design and target population. As shown in ► Figure 2 (A) and ► Figure 2 (B), the number of publications in terms of the study design and target population reflect the heterogeneity of using smart watches in healthcare. In the following discussion, we will examine the platform used, other related technologies, target population, usability testing, study design and their potential bias, and type of sensors used.

Based on our review, the platform most commonly used in healthcare research was that of the Android Wear, and there was no research utilizing that of the Apple Watch. This is not surprising since the first Android Wear started shipping in July 2014, whereas the Apple Watch was not available until April 10, 2015. While our study was designed to review the literature on healthcare applications of smart watches, a large amount of selected studies utilized the combination of a smart watch and a smart phone [11, 15, 22, 25, 27, 29, 34]. Although the smart watch has emerged as a standalone computing device intended to be used by the wearers with or without the concomitant use of a smart phone, currently most smart watches rely on a smart phone to assist their computing or connection abilities. Perhaps because smart phones are so prevalent today, some researchers chose to conduct research based on the combination of a smart phone and a smart watch, or compare usage between the two. With the launching of the Apple Watch OS 2.0 and a later version having native apps support (that can run on the watch itself instead of the iPhone), and with the Android Wear, which can now work on its own with cellular support via 4G connectivity [35, 36], it is possible that wearable smart watches will become a reality for content providers and therefore an opportunity for healthcare applications.

One study used a multimodal approach, including a wrist worn smart watch, a Microsoft Kinect, and other devices, to act as an assistive technology for activity monitoring in the elderly [20]. Microsoft Kinect was developed for gaming purpose, however, developers have recognized that the motion sensing camera has potential for healthcare applications, due to its ability to track movements in three-dimensional (3D) space and to Kinect’s open software development kit [37]. In the literature, there are several studies that utilized Kinect to assist the diagnosis or monitoring of disease activity for movement disorders especially in PD [38–42]. A performance comparison of Kinect and smart watches demands further investigation.

Smart watches are being used as a platform for a variety of healthcare applications. Based on our review, the most common healthcare applications using smart watches focused on health monitoring or smart home environment for the elderly [11–12, 16, 20, 25–26]. Another major application is with chronically ill patients needing medication adherence monitoring [18, 27, 30]. This focus is particularly relevant since the United States is projected to experience rapid growth in its older population in the next four decades [43], which will increase demand for chronic care. According to a report released by Centers for Disease Control and Prevention (CDC), approximately 80% of older adults have one chronic condition, and 50% have at least two [44]. As seniors live longer, technology may become an indispensable aspect of modern life. There are a number of care issues related to seniors, individuals with disabilities, and their caregivers, which can potentially benefit from technology. Among them, fall detection and prevention, chronic disease management, and medication management are the leading three identified by the Aging Services Technology Study [45].

Fall detection for elderly adults has been playing an important role in smart home environment [46]. Thousands of research articles have been published in the literature, and a variety of products are available on the market for automatic fall monitoring. Although existing fall detection studies have been conducted with different sensor positions, the devices are usually placed on both the upper and lower body, and the most common device placement position is the waist [47]. With the advent of smart watches characterized by miniaturization and unobtrusiveness, wide application of fall detection algorithms in such devices are possible in the future. Nevertheless, use of wearable fall detection devices by older adults in real-world settings demands further research and improvement in accuracy [48].

Another category of research found on this review is related to smart watch applications in patients with neurologic diseases, including PD, Alzheimer‘s disease, epilepsy, and stroke [13, 15, 17, 19, 24, 29, 33, 34]. Neurologic diseases are amongst the major causes of disabilities, and those coping with these disabilities may benefit from assistive technology using smart watches. These studies used a variety of study designs and interventions utilizing smart watches, including those intended to help Alzheimer patients recognize familiar people, enable analysis and diagnosis of tremors, detect types of seizures in children and young adults, assist PD patients with voice and speech disorders, and assess symptoms and motor signs of PD. In the two ongoing clinical trials, researchers are testing the use of smart watches for monitoring activity feedback during in-patient stroke rehabilitation, and for monitoring physical activity (including falls and tremor) in PD patients [33–34]. In one of the larger clinical studies by Patterson [13] the use of a smartwatch to detect seizures had disappointing results, suggesting that while their use in laboratory settings holds promise, further development and evaluation in clinical settings are needed.

For assistive technologies to be successfully implemented into the current workflow, gaps between the design phase and user experience must be bridged. This is especially important in the case of smart watches, given their small screen size. Another focus from this review emphasizes the importance of enhancing the user experience through usability testing, to evaluate a product before implementation. However, only two studies utilized user-centered design in the design phase, and only one study described a user interface evaluation [15, 22, 27]. No studies followed rigorous usability testing guidelines [49]. Usability testing has been used to evaluate a variety of assistive devices, however, this testing often excluded individuals with disabilities [50]. Among the selected articles, two studies focused on groups of people with special needs, including patients with visual or hearing impairment [22–23]. Both of these studies utilized a combined smart watch – smart phone system. One aimed to develop a system for gesture control in assisting low vision people during daily life; the other was designed to identify the needs and expectations of deaf people related to using the smartwatch as an environmental sound alert. It will be important to consider user-centered design and usability testing in future trials.

Although most of the studies we identified focused on health monitoring and patients with chronic illnesses, one study aimed to help patients experiencing out-of-hospital cardiac arrest (OHCA). Gruenerbl et al. developed a Cardiopulmonary Resuscitation (CPR) feedback application for a smart watch, designed to allow untrained bystanders to perform CPR correctly in emergencies [21]. Using the accelerometer of the smart watch, a CPR application was developed to provide real time feedback during chest compression CPR with three screen-based feedback functionalities: frequency, depth, and counting. This study enrolled a total of 41 participants to perform CPR in manikins. Using the smart watch for assistance was significantly associated with increased rate and depth of chest compression, although the findings were not as promising in terms of high quality CPR [51]. The application developed by Gruenerbl and colleagues did provide a brand new concept of using smart watches to assist bystander CPR; however, it provided only on-screen reminders without audio and vibration feedback. Furthermore, there was no usability testing on the product.

In this review, more than half (13, 54%) of the selected studies adopted a quantitative approach by using experimental design of machine learning. Since most smart watches contain an accelerometer and a gyroscope, it is possible to utilize the motion detection sensors for different patient populations. As a form of artificial intelligence, machine learning involved the training of a computer based on data collected from prior examples [52]. For healthcare applications using smart watches via machine learning approaches, health related data can be collected and combined with appropriate algorithms to provide valuable results. Such data collecting process constitutes what Simon called “the sciences of the artificial” [53], and experimentation is the alternative way for learning algorithms to formalize complex analysis when theoretical evidence is lacking. As Langley wrote in his influential editorial entitled “Machine Learning as an Experimental Science” in the journal Machine Learning, an experiment involves systematically varying one or more independent variables and examining their effect on some dependent variables [54]. In order to improve the performance of dependent measures, a machine learning experiment requires a number of observations made under different conditions [55]. As shown in ► Figure 2 (B), motion detection using smart watches and machine learning can be found in a variety of healthcare applications including elderly health monitoring or smart home, food and diet monitoring, medication adherence monitoring, and movement disorders. Experiments have to be conducted to collect annotated datasets for training purposes. Based on our review, all selected articles rely on supervised machine learning algorithms for the tasks of classification or pattern recognition, and most studies chose N-fold cross validation. Threats to validity include small sample size, classifiers used, and lack of testing with alternative datasets.

Although a detailed discussion is beyond the scope of this review, there are a variety of factors that may affect performance measures in healthcare applications using smart watches and machine learning algorithms. In particular, the use of sensors and the related performance measures may be of interest to some of our readers. With respect to types of sensors used in the included studies, 67% of the studies used at least one sensor and 21% used the combination of an accelerometer and a gyroscope. An accelerometer is a sensor which measures acceleration in the 3D coordinate system and a gyroscope detects rotation. Theoretically, the combination use of both sensors can increase the accuracy of motion detection in a selected target population. Empirically, Alias et al. showed significant results using both gyroscope and accelerometer sensors with some filters in a stabilized and moving platform application [56]. Due to the heterogeneity of selected studies, however, there is currently insufficient evidence to draw any relevant conclusion regarding the performance of the combined sensors use. Expanded experimental studies are needed.

In sum, the impact of the smart watch in real world clinical practice or even emergency settings has yet to be determined. For smart watches to be commonly used in the clinical arena, researchers will need to adopt more rigorous study designs and conduct usability testing before full implementation of smart watch technologies into clinical settings.

5. Limitations

The smart watch is not a new concept. However, with the advent of Android Wear and Apple Watch it has attracted wide attention. Research articles regarding healthcare applications of smart watches are scarce, based on our search of the literature. In order to expand the range of our review, we searched all pertinent databases available, and we included studies presented in medical conferences, as well as ongoing clinical trials. In the search terms, we used smart watch or smartwatch as the main keywords to ensure a broader coverage of articles to be considered for inclusion. Due to the heterogeneous nature of different databases, the quality of the included studies varied greatly. Nevertheless, this review highlights that while there is potential for healthcare applications using smart watch technology, more rigorous studies of their use in clinical settings is needed.

6. Conclusion

Smart watches exhibit the advantages of small form factor and can be wrapped on the wrist for daily wear. Although the reported use of smart watch applications for patients with chronic diseases appear promising, we found only one study focused on managing patients in critical or emergency conditions. In order for these devices to gain wide acceptance by health professionals, rigorous research on their accuracy, completeness and effect on workflow should be conducted before smart watch applications are integrated into clinical practice. User studies to investigate ideal functionality, user interface design, and usability for a variety of clinical and patient settings are needed. Further research is required to understand the impact of smart watch applications on clinical practice.


The authors would like to thank Beryl Schulman, PhD, Jean O. Taylor, PhD, and Kristin Dew, MS for their assisting with preparation of the manuscript and critical review.



T.C.L. designed the systematic review and conducted initial database search; C.C.F. screened titles and abstracts of identified records; M.H.M.M. reviewed the selection; T.C.L. and C.M.F. reviewed the full text articles and contributed to data extraction. All authors participated in the drafting of the manuscript and review of the content. A.M.T. supervised the whole process of the systematic review. All authors approved the final version of the review.

Clinical Relevance Statement

With appropriate design and rigorous research, smart watches have the potential to improve many aspects of healthcare delivery.

Conflict of interest

The authors declare that they have no conflicts of interest in the research.

Human Subjects Protections

Human and/or animal subjects were not included in the project.


1. Patel MS, Asch DA, Volpp KG. Wearable devices as facilitators, not drivers, of health behavior change. JAMA 2015; 313: 459-460. [PubMed]
2. CCS Insight. Wearables Market to Be Worth $25 Billion by 2019. [cited 2016 May 17]. Available from
3. Lukowicz P, Kirstein T, Tröster G. Wearable systems for health care applications. Methods Inf Med 2004; 43: 232-238. [PubMed]
4. Ajami S, Teimouri F. Features and application of wearable biosensors in medical care. J Res Med Sci 2015; 20: 1208-1215. [PMC free article] [PubMed]
5. Schroetter J. The Future of Wearable Computing in Healthcare. 2014. August [cited 2015 Dec 3]. Available from
6. Poon CC, Zhang YT. Perspectives on high technologies for low-cost healthcare. IEEE Eng Med Biol Mag 2008; 27: 42-47. [PubMed]
7. Zheng YL, Ding XR, Poon CC, Lo BP, Zhang H, Zhou XL, Yang GZ, Zhao N, Zhang YT. Unobtrusive sensing and wearable devices for health informatics. IEEE Trans Biomed Eng 2014; 61: 1538-1554. [PubMed]
8. O‘Donnell B. Smartwatches: The New Smartphones Jr.? 2015 Apr; [cited 2015 Dec 3]. Available from
9. Scher DL. Will the Apple Watch Revolutionize Healthcare? Medscape Business of Medicine. 2015. June [cited 2015 Dec 3]. Available from:
10. Moher D, Liberati A, Tetzlaff J, Altman DG.; PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ 2009; 339: b2535. [PubMed]
11. Casilari E, Oviedo-Jiménez MA. Automatic Fall Detection System Based on the Combined Use of a Smart-phone and a Smartwatch. PLoS One 2015; 10: e0140929. [PMC free article] [PubMed]
12. Mortazavi B, Nemati E, VanderWall K, Flores-Rodriguez HG, Cai JY, Lucier J, Naeim A, Sarrafzadeh M. Can Smartwatches Replace Smartphones for Posture Tracking? Sensors (Basel) 2015; 15: 26783-26800. [PMC free article] [PubMed]
13. Patterson AL, Mudigoudar B, Fulton S, McGregor A, Poppel K V, Wheless MC, Brooks L, Wheless JW. SmartWatch by SmartMonitor: Assessment of Seizure Detection Efficacy for Various Seizure Types in Children, a Large Prospective Single-Center Study. Pediatr Neurol 2015; 53: 309-311. [PubMed]
14. Kalantarian H, Sarrafzadeh M. Audio-based detection and evaluation of eating behavior using the smart-watch platform. Comput Biol Med 2015; 65: 1–9. [PubMed]
15. Fardoun HM, Mashat AA, Ramirez Castillo J. Recognition of familiar people with a mobile cloud architecture for Alzheimer patients. Disabil Rehabil 2015: 1–5. [PubMed]
16. Carlson JD, Mittek M, Parkison SA, Sathler P, Bayne D, Psota ET, Perez LC, Bonasera SJ. Smart watch RSSI localization and refinement for behavioral classification using laser-SLAM for mapping and fingerprinting. Conf Proc IEEE Eng Med Biol Soc 2014; 2014: 2173-2176. [PubMed]
17. Wile DJ, Ranawaya R, Kiss ZH. Smart watch accelerometry for analysis and diagnosis of tremor. J Neurosci Methods 2014; 230: 1-4. [PubMed]
18. Sailer F, Pobiruchin M, Wiesner M. An Approach to Improve Medication Adherence by Smart Watches. In the 26th Medical Informatics Europe Conference: Digital Healthcare Empowering Europeans: Proceedings of MIE2015; 2015. May 27–29; Madrid, Spain.
19. Gazit E, BernadElazari H, Moore ST, Cho C, Kubota K, Vincent L, Cohen S, Reitblat L, Fixler N, Mirelman A, Giladi N, Hausdorff JM. Assessment of Parkinsonian motor symptoms using a continuously worn smartwatch: Preliminary experience. Mov Disord 2015; 30 (Suppl. 1): S272.
20. Ahanathapillai V, Amor JD, James CJ. Assistive technology to monitor activity, health and wellbeing in old age: The wrist wearable unit in the USEFIL project. Technol Disabil 2015; 27: 17-29.
21. Gruenerbl A, Prikl G, Monger E, Gobbi M, Lukowicz P. Smart-watch life saver: Smart-watch interactive-feedback system for improving bystander CPR. In The 19th International Symposium on Wearable Computers (ISWC 2015); 2015. September 7–11; Osaka, Japan. New York: ACM, 2015.
22. Porzi L, Messelodi S, Modena CM, Ricci EA. smart watch-based gesture recognition system for assisting people with visual impairments. In Proceedings of the 3rd ACM international workshop on Interactive multimedia on mobile & portable devices, pp. 19–24; 2013. October 21–25; Barcelona, Spain. New York: ACM, 2013.
23. Mielke M, Brück R. A Pilot Study about the Smartwatch as Assistive Device for Deaf People. In Proceedings of the 17th International ACM SIGACCESS Conference on Computers & Accessibility, pp. 301–2; 2015 Oct 26–28; Lisbon, Portugal. New York: ACM, 2015.
24. Dubey H, Goldberg JC, Abtahi M, Mahler L, Mankodiya K. EchoWear: Smartwatch Technology for Voice and Speech Treatments of Patients with Parkinson‘s Disease. In Proceedings of the conference on Wireless Health; 2015 Oct 14–16; Bethesda, MD, USA. New York: ACM, 2015.
25. Lee S, Kim Y, Ahn D, Ha R, Lee K, Cha H. Non-obstructive Room-level Locating System in Home Environments using Activity Fingerprints from Smartwatch. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing, pp. 939–50; 2015 Sep 7–11; Osaka, Japan. New York; ACM, 2015.
26. Thomaz E, Essa I, Abowd GD. A Practical Approach for Recognizing Eating Moments with Wrist-Mounted Inertial Sensing. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing, pp.1029–40; 2015 Sep 7–11; Osaka, Japan. New York; ACM, 2015.
27. Maglogiannis I, Spyroglou G, Panagopoulos C, Mazonaki M. Mobile reminder system for furthering patient adherence utilizing commodity smartwatch and Android devices. In 2014 EAI 4th International Conference on Wireless Mobile Communication and Healthcare (Mobihealth), pp. 124–7; 2014 Nov 3–5; Athens, Greece: IEEE Xplore, 2014.
28. Sen S, Subbaraju V, Misra A, Balan RK, Lee Y. The case for smartwatch-based diet monitoring. In 2015 IEEE International Conference on Pervasive Computing and Communication Workshops (PerCom Workshops), pp.585–90; 2015 Mar 23–27; St. Louis, Missouri, USA: IEEE Xplore, 2015.
29. Sanders TH, Clements MA. Multimodal monitoring for neurological disorders. In 2014 40th Annual Northeast on Bioengineering Conference; 2014 Apr 25–27; Boston, Mass., USA: IEEE Xplore, 2014.
30. Kalantarian H, Alshurafa N, Nemati E, Le T, Sarrafzadeh M. A Smartwatch-Based Medication Adherence System. In 2015 IEEE 12th International Conference on Wearable and Implantable Body Sensor Networks; 2015 Jun 9–12; Cambridge, MA, USA: IEEE Xplore, 2015.
31. Jovanov E. Preliminary Analysis of the Use of Smartwatches for Longitudinal Health Monitoring. In 2015 37th Annual International Conference of the IEEE on Engineering in Medicine and Biology Society; 2015 Aug 25–29; Milan, Italy: IEEE Xplore, 2015. [PubMed]
32. Ye X, Chen G, Cao Y. Automatic Eating Detection Using Head-Mount and Wrist-Worn Accelerometers. In 2015 17th International Conference on E-health Networking, Application & Services (HealthCom). 2015 Oct 14–17; Boston, MA, USA: IEEE Xplore, 2015.
33. Steins D. The Effect of Activity Feedback Enabled by Smart Watches During In-patient Stroke Rehabilitation. 2015. October [cited 2015 Dec 3]; Available from:
34. Faber MJ. RealPD Trial: Development of Clinical Prognostic Models for Parkinson‘s Disease. 2015. August [cited 2015 Dec 3]; Available from:
35. Apple Inc. Developing for watchOS. [cited 2015 Dec 3]; Available from:
36. Android Official Blog. Cellular support comes to Android Wear. [cited 2015 Dec 3]; Available from:
37. Comstock J. Eight ways the Microsoft Kinect will change healthcare. [cited 2016 May 20]; Available from:
38. Álvarez M, Bosch J, Martínez A, Macías F, Valdéz P. 3D sensors, the new paradigm for assessing movement disorders. In the MDS 17th International Congress of Parkinson‘s Disease and Movement Disorders: Movement Disorder; 2013. June 16-20; Sydney, Australia.
39. Galna B, Barry G, Jackson D, Mhiripiri D, Olivier P, Rochester L. Accuracy of the Microsoft Kinect sensor for measuring movement in people with Parkinson‘s disease. Gait Posture 2014; 39: 1062-1068. [PubMed]
40. Ťupa O, Procházka A, Vyšata O, Schätz M, Mareš J, Vališ M, Mařík V. Motion tracking and gait feature estimation for recognising Parkinson‘s disease using MS Kinect. Biomed Eng Online 2015; 14: 97. [PMC free article] [PubMed]
41. Torres R, Huerta M, Clotet R, González R, Sagbay G, Erazo M, Pirrone J. Diagnosis of the corporal movement in Parkinson’s Disease using Kinect Sensors. In the World Congress on Medical Physics and Biomedical Engineering: IFMBE Proceedings; 2015. June 7-12, Toronto, Canada.
42. Elgendi M, Picon F, Magnenat-Thalmann N, Abbott D. Arm movement speed assessment via a Kinect camera: A preliminary study in healthy subjects. Biomed Eng Online 2014; 13: 88. [PMC free article] [PubMed]
43. Vincent GK, Velkoff VA. The next four decades, the older population in the United States: 2010 to. 2050, Current Population Reports. Washington, DC: U.S. Census Bureau; 2010: 25–1138.
44. Healthy aging; helping people to live long and productive lives and enjoy a good quality of life. At a glance 2011. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Division of Adult and Community Health; [Internet]. [cited 2015 Dec 3]; Available from:
45. Report to Congress: Aging services technology study. Washington, DC: U.S. Department of Health and Human Services, Assistant Secretary for Planning and Evaluation, Office of Disability, Aging and Long-Term Care Policy; 2012. June [cited 2015 Dec 3]; Available from:
46. Rakhman AZ, Kurnianingsihi Nugrohoi LE., Widyawani. u-FASt: Ubiquitous fall detection and alert system for elderly people in smart home environment, in 2014 Electrical Engineering and Informatics (MICEEI) on Makassar International Conference, pp.136–140; 2014. November 26–30; Makassar, Indonesia: IEEE Xplore; 2014.
47. Pannurat N, Thiemjarus S, Nantajeewarawat E. Automatic fall monitoring: a review. Sensors (Basel). 2014;14:12900–36. [PMC free article] [PubMed]
48. Chaudhuri S, Oudejans D, Thompson HJ, Demiris G. Real-World Accuracy and Use of a Wearable Fall Detection Device by Older Adults. J Am Geriatr Soc 2015; 63: 2415-2416. [PMC free article] [PubMed]
49. Rubin J, Chisnell D. Handbook of usability testing: how to plan, design, and conduct effective tests, 2nd ed Indianapolis: Wiley Publishing; 2008.
50. Quesenbery W. Usable Accessibility: Making Web Sites Work Well for People with Disabilities. 2009. February [cited 2015 Dec 3]; Available from:
51. Meaney PA, Bobrow BJ, Mancini ME, Christenson J, de Caen AR, Bhanji F, Abella BS, Kleinman ME, Edelson DP, Berg RA, Aufderheide T P, Menon V, Leary M.; CPR Quality Summit Investigators the American Heart Association Emergency Cardiovascular Care Committee the Council on Cardiopulmonary, Critical Care, Perioperative and Resuscitation. Cardiopulmonary resuscitation quality: [corrected] improving cardiac resuscitation outcomes both inside and outside the hospital: a consensus statement from the American Heart Association. Circulation 2013; 128: 417-435. [PubMed]
52. Jain K. Machine Learning basics for a newbie. 2015. June [cited 2016 Jun 13]. Available from
53. Simon HA. The Sciences of the Artificial. Cambridge, MA: MIT press; 1969.
54. Langley P. Machine learning as an experimental science. Mach Learn. 1988: 3: 5–8.
55. Langley P, Kibler D. The experimental study of machine learning. NASA Ames Research Center, Moffett Field, CA (1991). Unpublished Report. [cited 2016 May 17]. Available from
56. Alias AR, Alias MS, Shamsuddin IZ, Ahmad RAR, Abdullah SNHS. Measure the Ability and Limitation of Gyroscope, Acceleration and Gyro-accelaration for Stabilized Platform. In the16th FIRA RoboWorld Congress: FIRA 2013 Proceedings; 2013. August 24–29; Kuala Lumpur, Malaysia.

Articles from Applied Clinical Informatics are provided here courtesy of Thieme Medical Publishers