J Med Syst (2015) 39:174 DOI 10.1007/s10916-014-0174-2

MOBILE SYSTEMS

Decision Support Systems and Applications in Ophthalmology: Literature and Commercial Review Focused on Mobile Apps Isabel de la Torre-Díez & Borja Martínez-Pérez & Miguel López-Coronado & Javier Rodríguez Díaz & Miguel Maldonado López

Received: 16 September 2014 / Accepted: 25 November 2014 # Springer Science+Business Media New York 2014

Abstract The growing importance that mobile devices have in daily life has also reached health care and medicine. This is making the paradigm of health care change and the concept of mHealth or mobile health more relevant, whose main essence is the apps. This new reality makes it possible for doctors who are not specialist to have easy access to all the information generated in different corners of the world, making them potential keepers of that knowledge. However, the new daily information exceeds the limits of the human intellect, making Decision Support Systems (DSS) necessary for helping doctors to diagnose diseases and also help them to decide the attitude that has to be taken towards these diagnoses. These could improve the health care in remote areas and developing countries. All of this is even more important in diseases that are more prevalent in primary care and that directly affect the people’s quality of life, this is the case in ophthalmological problems where in

This article is part of the Topical Collection on Mobile Systems I. de la Torre-Díez (*) : B. Martínez-Pérez : M. López-Coronado : J. R. Díaz Department of Signal Theory and Communications, and Telematics Engineering, University of Valladolid, Paseo de Belén, 15, 47011 Valladolid, Spain e-mail: [email protected]

first patient care a specialist in ophthalmology is not involved. The goal of this paper is to analyse the state of the art of DSS in Ophthalmology. Many of them focused on diseases affecting the eye’s posterior pole. For achieving the main purpose of this research work, a literature review and commercial apps analysis will be done. The used databases and systems will be IEEE Xplore, Web of Science (WoS), Scopus, and PubMed. The search is limited to articles published from 2000 until now. Later, different Mobile Decision Support System (MDSS) in Ophthalmology will be analyzed in the virtual stores for Android and iOS. 37 articles were selected according their thematic (posterior pole, anterior pole, Electronic Health Records (EHRs), cloud, data mining, algorithms and structures for DSS, and other) from a total of 600 found in the above cited databases. Very few mobile apps were found in the different stores. It can be concluded that almost all existing mobile apps are focused on the eye’s posterior pole. Among them, the most intended are for diagnostic of diabetic retinopathy. The primary market niche of the commercial apps is the general physicians.

Keywords Decision Support System (DSS) . Eye’s posterior pole . mHealth . Ophthalmology

B. Martínez-Pérez e-mail: [email protected] M. López-Coronado e-mail: [email protected] J. R. Díaz e-mail: [email protected] M. M. López University Institute of Applied Ophthalmobiology (IOBA), University of Valladolid, Paseo de Belén, 17. Campus Miguel Delibes, 47011 Valladolid, Spain

Abbreviations CDS Clinical Decision Support CDSS Clinical Decision Support System DSS Decision Support System EHR Electronic Health Record MDSS Mobile Decision Support System SOA Service-Oriented Architecture WoS Web of Science

174, Page 2 of 10

Introduction Medical knowledge in modern healthcare is constantly changing. In this context, computer programs provide clinical decisions support to physicians and help them in the fight against incorrect and / or incomplete diagnosis. This task is performed by the Decision Support Systems (DSS). They provide physicians, patients, staff or other individuals with knowledge and person-specific information, intelligently filtered or presented at appropriate times, to enhance health. Moreover, they encompass a variety of tools to enhance decision-making in the clinical workflow. Some of these tools include computerized alerts or clinical guidelines, focused on patient data reports, among others [1]. These systems also improve the usability of the traditional methods. An analysis developed by [2] showed that the usability problems experienced by health professionals when a guide on paper is used could be overcome by applying usercentric Mobile Decision Support System (MDSS). However, the existing MDSS have hardly been applied in a practical way [3] and the enormous potential they have to optimize the health care system is far from being reached [1]. This great potential is beginning to be exploited by smartphones that are creating new values in the domains of health, to the point of being considered powerful platforms to provide affordable solutions and approaches to generalized scales to the problems posed by the patients [4]. According to industry estimates, 500 million smartphone users worldwide will be using an app for health care in 2015 and, by 2018, over 3.4 billion users of mobile devices, including professionals and patients, will have downloaded a MDSS [5]. If we commented that generally mobile DSS are just exploiting their enormous potential, ophthalmology is not pulled from this script. Martinez-Pérez et al. (2014) [6] conducted a study of existing applications for DSS and concluded that this is one of the fields of medicine with fewer systems of this type. Besides, of the existing ones, most of them are focused on the eye’s posterior pole and more particularly in the early diagnosis of diabetic retinopathy using complex devices for image analysis. Among its conclusions, the significant growth in the number of mobile systems for decision support in the past 2 years is highlighted. With this, developers are proposed to make applications in the fields of medicine with fewer systems of this type, included ophthalmology. Despite the shortage of DSS in the field of ophthalmology, ophthalmic motives are usually one of the most common reasons for medical consultation and are especially important because the first does not involve specialists. On the other hand, from eye care consultations, the red eye is a considerable concern in the patient and family and is presented as one of the most important in quantitative terms, although in most cases is due to not very serious causes. But the red eye can be caused by diseases that threaten vision or the integrity of the

J Med Syst (2015) 39:174

eye. It is therefore very important that the general practitioner knows the most common causes and has a definite criterion to decide between handle it himself or refer the case to an ophthalmologist [7]. The main aim of this paper is to analyze the state of the art focused on DSS in Ophthalmology. This study will lay the groundwork for a future mobile DSS app for Android devices directed to staff not specialized in diseases of the eye’s anterior pole. The remainder of this paper is organized as follows. Next section shows the methodology conducted in this review and then there are shown the results obtained. Both sections are separated into two subsections: the first subsection is the literature review of systems, and the second is the review of commercial apps. Finally, the last section presents the discussion of the results.

Methodology As mentioned in the Introduction section, Martínez-Pérez et al. (2014) [6] conducted a study of existing mobile applications of DSS in order to find the different apps in this field, compare common features and analyze a representative sample of them. In this work it was decided to analyze the DSS applied to ophthalmology proposed so far. To do this, both the literary part and the business side will be studied. That is, several academic databases and search systems will be checked first. These are: IEEE Xplore [8], Web of Science (WoS) [9], Scopus [10] and PubMed [11]. Subsequently, the commercial applications of MDSS focused in ophthalmology that are available on the two more widespread operating systems for smartphones, Android and iOS, will be analyzed. Methods for the literature review The terms combination used for the search of relevant papers in the mentioned databases is the following: (DSS OR CDSS OR “Decision Support” OR “Diagnosis Support”) AND (eye OR ophthalmology). Furthermore, the search is limited to articles published from 2000 onwards. Importantly, the search order is as indicated (IEEE Xplore, Web of Science, Scopus and PubMed), as this will be especially relevant when select the items found. To select a paper from the returned results its title is analyzed, and in cases of doubt, the abstract or the full article is read if necessary. Similarly, those items listed as recommended or related are analyzed. Subsequently, the selected items are classified into two groups: (i) either by itself be the content interesting or be strongly related to the topic treated or by (ii) contain certain aspects that may be interesting for the development of the project. This classification is done

J Med Syst (2015) 39:174

individually for all databases and later raised all together to have a more comprehensive view of the results of the review. Finally, among the selected items, we proceed to classify those who are directly applied to ophthalmology, and those which, without treat eye issues directly, are related to the topic of the project. Those related to ophthalmology are further divided into: (i) those dealing with diseases of the posterior pole of the eye, (ii) dealing with the anterior pole and (iii) those that have a more generic content. Meanwhile, those papers that not directly treat eye issues are classified into: (i) those dealing with the structure and main algorithms used in the DSS, (ii) those who talks about the Electronic Health Record (EHR) or storage of medical data in the cloud or data mining applied to these systems, and (iii) those that cannot be grouped into any of the mentioned classifications. Methods for searching commercial apps The search of commercial applications related to the topic of this project was conducted in the stores of the two most widespread operating systems for smartphones, Google Play [12] for Android and the App Store [13] for iOS. The search words used in both stores were “DSS”, “CDSS”, “ophthalmology”, “eye”, “oftalmología” and “ojo.” It is important to note that, unlike the case in the literature review, in this case it is not possible to use combinations of words, being only available the option of searching for words or sets of words without logical operators (AND, OR,…). To select an app from the returned results, its title and its topic is analyzed first, considering only those that are developed in English or Spanish. Of those selected, their description are read and, in case of doubt and if the app is free, it is tested on a device to analyze and see its usefulness. To do this, a Samsung Galaxy S4 for Android and an iPhone 5S for iOS are used. Fig. 1 Summary flowcharts followed in the reviews of the set of databases used

Page 3 of 10, 174

It is noteworthy that most of the findings were related to ophthalmology congresses, atlas of eye diseases, visual acuity test or typical exam questions on ophthalmology subjects. By contrast, just few results of DSS were found.

Results Literature review In this section the results obtained from the literature review raised above are presented. Figure 1 shows a flow chart with the results obtained in the set of databases (IEEE Xplore, Web of Science, Scopus and PubMed). In Fig. 2 a new classification of the 37 chosen papers is shown, depending on their subject matter and their importance within the framework of this paper. In the following sections the main findings presented in the articles of each block are explained and it will be presented, in each of them, a table of selected articles sorted by year of publication indicating its title, authors, year of publication and journal in which were published. Eye’s posterior pole Table 1 shows a list of items whose most significant issue is the back of the eye. In the 15 papers selected in this section and shown in Table 1 and in the other papers related to the posterior pole of the eye, it can be seen that almost all of them are focused on the early detection of diabetic retinopathy using the image analysis. So, in 2004, Kahai P. et al. (2004) [3] proposed a system for early detection of this disease. Later, in 2006, Paunksnis P. et al. [26] stated in their article that it may be possible early

174, Page 4 of 10

J Med Syst (2015) 39:174

Fig. 2 Classification of the selected papers in the search

detection of glaucoma if the search patterns are simplified. Following this, numerous projects applying different techniques to improve DSS arise: in [25] using data mining is proposed to distinguish the images of diseased eyes from healthy eyes. In [23] and [24], support systems for medical Table 1

decision based on the internet that allows distributed processing and storage of the images of the eye to create a database of specialized medical data is presented; besides, in [23] signals used to detect diseases of the retina and fundus are also used to detect heart disease. Similarly, in [21] a database with type I

Some of the papers selected in Ophthalmology: posterior pole

Title

Authors

Year Journal

An intelligent mobile based decision support system for retinal disease diagnosis [4] Decision Support System for Detection of Diabetic Retinopathy Using Smartphones [14] Retinal image analysis aimed at extraction of vascular structure using linear discriminant classifier [15] Computer-aided diagnosis of diabetic retinopathy: A review [16]

Bourouis A, Feham M, Hossain MA, Zhang L.

2014 Scopus

Prasanna P, Jain S, Bhagatt N, Madabhushi A.

2013 IEEE

Fraz MM, Remagnino P, Hoppe A, Barman SA.

2013 IEEE

Decision support system for diabetic retinopathy using discrete wavelet transform [17] Telemedicine and ocular health in diabetes mellitus [18] An improved medical decision support system to identify the diabetic retinopathy using fundus images [19] Retinal image registration and comparison for clinical decision support [20] A hybrid Decision Support System for the Risk Assessment of retinopathy development as a long term complication of Type 1 Diabetes Mellitus [21] Support system for the preventive diagnosis of Hypertensive Retinopathy [22] Network Based Clinical Decision Support System [23]

Mookiah MRK, Rajendra U, Kuang C, Min C, Ng EYK, Laude 2013 Scopus A. Noronha K, Acharya UR, Nayak KP, Kamath S, Bhandary SV. 2012 WoS Bursell SE, Brazionis L, Jenkins A. Kumar SJ, Madheswaran M.

2012 PubMed 2012 PubMed

Xiao D, Vignarajan J, Lock J, Frost S, Tay-Kearney ML, Kanagasingam Y. Skevofilakas M, Zarkogianni K, Karamanos BG, Nikita KS.

2012 PubMed

Ortíz D, Cubides M, Suárez A, Zequera M, Quiroga J, Gómez J, Arroyo N. Jegelevicius D, Krisciukaitis A, Lukosevicius A, Marozas V, Paunksnis A, Barzdziukas V, Patasius M, Buteikiene D, Vainoras A, Gargasas L. Automated Retinal Image Analysis Over the Internet [24] Chia-Ling T, Madore B, Leotta MJ, Sofka M, Gehua Y, Majerovics A, Tanenbaum HL, Stewart CV, Roysam B. Spatial Modeling and Classification of Corneal Shape [25] Marsolo K, Twa M, Bullimore MA, Parthasarathy S. The use of information technologies for diagnosis in ophthalmology Paunksnis A, Barzdziukas V, Jegelevicius D, Kurapkiene S, [26] Dzemyda G. Decision Support for Automated Screening of Diabetic Retinopathy Kahai P, Namuduri KR, Thompson H. [3]

2010 IEEE

2010 IEEE 2009 IEEE

2008 IEEE 2007 IEEE 2006 PubMed 2004 IEEE

J Med Syst (2015) 39:174 Table 2

Page 5 of 10, 174

Anterior pole: Selected Articles in Ophthalmology

Title

Authors

Year

Generic apps Journal

Lepo: Sistema de Apoio à Roque AC, Mantovani J, 2004 IEEE Decisão Médica em Torres I, Galina AC, Oftalmologia Baseado no de Lima PR, Sistema Lepidus [27] Novoa C, Schor P.

diabetes patients is used to predict the risk of these patients of suffering from diabetic retinopathy. It is important to note that the mentioned article does not make use of an electronic clinical record, but could be considered similar in content. [17–20] used different methods to classify fundus images and extract analysis parameters to predict the probability that the patient has diabetic retinopathy. In recent years the trend is changing. While prior systems for early detection of diabetic retinopathy were focused on the development of algorithms, in recent years there is special emphasis on its implementation. Thus, [4, 14] propose concrete support systems for medical decision intended to detect diabetic retinopathy using Android smartphones. These devices are intended for ophthalmologists, general practitioners and emergency physicians and have a small gadget that is incorporated into the phone and works as a lens or a ophthalmoscope and allows the physician to see the patient’s fundus.

Eye’s anterior pole Table 2 shows the only paper from the selected whose main theme is the anterior pole of the eye. The above article, developed by a university in Brazil, presents a specialization in the field of ophthalmology of a general medicine system that intends to help doctors in diagnosing diseases. It is based on the use of damped sinusoidal functions to represent each disease so that each feature of the disease affects a parameter of the function. Subsequently, the resulting sinusoidal signal is compared with those stored in the database and a list of possible diseases that the patient may suffer is offered.

Table 3

Table 3 shows a list of articles focusing on ophthalmology that do not focus on the anterior pole or in the anterior pole of the eye. Some of the items shown in Table 3, like most of Table 1, are based on the analysis of images to assist the medical decision. This is the case of [29] and [31]; the first analyzes the images to help diagnose the most typical eye diseases of the elderly. The second analyzes the movements of the eyes to detect eye abnormalities. In [30], the authors take into account the patient’s history as a tool to facilitate the diagnosis of eye disease. This method can be helpful due to the use of the EHRs, and even more if the conclusions of [28] are taking into account, which states that the satisfaction of ophthalmologists using an EHR is high. Structures and algorithms (DSS) The articles in this section (see Table 4) try to deal with the big problems that people who want to develop DSS face, both from the point of view of design and from construction. Thus, in [1] the main difficulties that arise when you want to build a help system (large amount of data including variables and interrelated, databases, models, site access, etc.) are presented and claims at the date of its publication (2001) that it is essential to have more powerful infrastructures than in any other work environment, thereby presenting a framework that allows to address the above difficulties. This article also highlights the importance of the interface for the acceptance of the product. Finally the authors note that medical informatics has grown to be considered a new discipline; this growth is mainly due to increased use of technology, population mobility, and specialization, use of computer systems, increased costs, improvements in hardware and improvements in methods. [34] presents another problem which must be faced, and it is the difficulties of understanding between the developer and the expert user. These problems are largely due to differences in training, experience and expectations, which lead to errors in the information-knowledge transfer from one party to

Some of the items selected in Ophthalmology: Generic

Title

Authors

Year

Journal

Adoption and Perceptions of Electronic Health Record Systems by Ophthalmologists: An American Academy of Ophthalmology Survey [28] A Computer-based classification of eye diseases [29]

Chiang MF, Boland MV, Margolis JW, Lum F, Abramoff MD, Hildebrand PL. Acharya UR, Kannathal N, Ng EY, Min LC, Suri JS. Odufuwa T, Bola O, Solebo L, Low S. Xiao-Peng H, Dempere L, Yang GZ.

2008

WoS

2006

PubMed

2006 2003

WoS IEEE

Diagnostic decision support in ophthalmology [30] Hot spot detection based on feature space representation of visual search in medical imaging [31]

174, Page 6 of 10 Table 4

J Med Syst (2015) 39:174

Selected items within Other: Structures and algorithms DSS

Title

Authors

Year

Journal

Acceptability and Difficulties of (Fuzzy) Decision Support Systems in Clinical Practice [1] Application of probabilistic and fuzzy cognitive approaches in semantic web framework for medical decision support [32] Systematic causal knowledge acquisition using FCM Constructor for product design decision support [33] Preventing knowledge transfer errors: Probabilistic decision support systems through the users’ eyes [34]

Schuh CJ, de Bruin JS. Seeling W.

2013

IEEE

Papageorgiou EI, Huszka C, De Roo J, Douali N, Jaulent MC, Colaert D. Ping W, Kimb S, Kimb KY, Yangc HJ.

2013

WoS

2011

WoS

Tabachneck-Schijf HJM, Geenen PL.

2009

WoS

another. To partially solve this problem, the authors propose a user-centered design and present serious attention to the translation of the terms and the knowledge transfer. At the edge of this problem, [32, 33] present a semantic web that represents medical knowledge and reasoning processes used in the design of support systems for medical decision. EHRs, cloud and data mining In this section is presented the influence and the great potential that electronic medical records, data mining and cloud storage (Table 5) may have in DSS. More recently (2014) in [35] the opinion of several American doctors who have worked several years with EHR is analyzed and the possibility of improving the DSS is stressed. In 2013, Dixon B. et al. [36] conclude that the decision support in the cloud is feasible and can be a very reasonable way to achieve better support in making clinical decisions. Finally [39] and [40] discuss the potential importance of data mining applied to the DSS. Define data mining as a technique for obtaining information that is hidden to the human eye and can help save lives in the field of DSS. The disadvantages of data mining are privacy, security and the misuse of information, issues that are of vital importance in the clinical data. Finally EHR are related to mining, stating that the application of data mining to EHR can help recover Table 5 Title

undiscovered knowledge and thereby improving clinical diagnoses. Other apps In this section those papers relevant to the topic of the project that cannot be embedded within the other sections (see Table 6) are included. Notable among them is the article of Martínez-Pérez et al. (2014) [6], the basis of this paper. On the other hand, [41] presents the creation of a visual dictionary with eye disease, which has a similar approach to the teaching part of the present project. In [2] the usability of SAD is analyzed in comparison with the paper guides claiming that usability problems experienced by health professionals when a guide is used on paper could be overcome by applying a user-centered design guideline of CDSS. These findings again highlight the broad acceptance that these systems can have among health professionals. Finally [42] presents a similar approach to this project, but instead of focusing on eye diseases, focusing on diabetes. It begins highlighting the serious effects this disease can have and the importance of an early diagnosis. The system asks the user questions about the signs, symptoms and risk factors and the user must give a “yes or no” answer. From these answers, the DSS gives a presumptive diagnosis of a disease, the probability of suffering and the severity of it. Other authors,

Selected items within Other: EHR, Cloud and Data Mining Authors

The good, the bad and the early adopters: providers’ attitudes about Makam AN, Lanham HJ, Batchelor K, Moran B, Howella common, commercial EHR [35] Stampley T, Kirk L, Cherukuri M, Samal L, Santini N, Leykum LK, Halm EA. A pilot study of distributed knowledge management and clinical Dixon B, Simonaitis L, Goldberg H, Paterno M, Schaeffer M, decision support in the cloud [36] Hongsermeier T, Wright A, Middleton B. Applying Data Mining Techniques to Standardized Electronic Batra S, Parashar HJ, Sachdeva S, Mehndiratta P. Health Records for Decision Support [37] Three-level HAC on food borne disease and related treatment to Kadam T, Chitre V. help medical DSS [38] E-Healthcare and Data Management Services in a Cloud [39] Ahmed S, Abdullah A. Design and implementation of a standards-based interoperable InSook C, JeongAh K, JiHyun K, Hyun YK, Yoon K. clinical decision support architecture in the context of the Korean EHR [40]

Year Journal 2014 WoS

2013 WoS 2013 IEEE 2012 IEEE 2011 IEEE 2010 WoS

J Med Syst (2015) 39:174 Table 6

Page 7 of 10, 174

Selected items within Other: Other

Title

Authors

Year

Journal

Mobile Clinical Decision Support Systems and Applications: A Literature and Commercial Review [6]

Martínez-Pérez B, de la Torre-Díez I, López-Coronado M, Sainz de Abajo B, Robles M, García-Gómez JM. Madhukumar S, Vijayalakshmi R.

2014

PubMed

2013

IEEE

2013

WoS

2012

Scopus

Visual dictionary: A decision support tool for DR pathology detection on POI [41] From an expert-driven paper guideline to a user-centred decision support system: A usability comparison study [2] Diabetes diagnosis decision support system based on symptoms, signs and risk factor using special computational algorithm by rule base [42]

as Rodríguez Loya et al. (2014) propose a Service-Oriented Architecture (SOA) by facilitating reliable Clinical Decision Support (CDS) [43].

Commercial apps In this section the results obtained from the review of commercial applications are presented. Some of the applications that have been selected are shown with its main task and some screenshots. “Wills Eye Manual”. This is the only selected app within the field of ophthalmology in the analysis done by [6]. It’s an app for the diagnosis and monitoring of optical diseases that features two flowcharts of decision support. This app is not tested on any device because it has a cost of 72 €. “Eye Conditions & Treatments”. It is a guide in English for optical diseases. For each disease indicates the conditions, causes, diagnosis and treatment. Its price is 2 €. “Oftalmología”. It’s a free app in Spanish to identify the different diseases that can occur in the fundus. Therefore it is not a DSS, but rather a guide to diseases. It is also important to note that this application is intended for diseases of the fundus and therefore it does not cover the scope of the present project (posterior pole of the eye). “Ophthalmology - Medical Dict” and “OphthalmologyDictionary”. They are two examples of dictionaries of terms related to ophthalmology. Both are free and are developed in English. “EClinician” and “Isabel”. According to their description, both are DSS for general medicine. They have a similar structure as the application developed in line with this project, with selection menus and giving importance to the clinical images. “Eye Handbook”. It is an application that has several features related to ophthalmology. First, it has a forum and allows participants to see the profile of users, as long as you are registered. It also has a guide with images of disease and has the option of displaying an image, ask what disease is treated and then give you the answer and diagnosis. Finally, it

Kilsdonk E, Peute LW, Riezebos RJ, Kremer LC, Jaspers M. Rahaman S.

provides links to various wikis that treat eye issues, to videos and related external pages, and also to visual acuity test. “Ophthalmology”. This is an application in English to medicine students. When initializing the app 4 menus (Fig. 3) appear: eye diseases, eye examination, test and evaluate the app. Entering in eye diseases menu, different submenus appear: differential diagnosis, ocular emergency, optics, orbital, lacrimal apparatus, eyelids and eyelashes, conjunctiva, sphere, cornea etc. Each one of these has a new classification, e.g. differential diagnosis menu has vision loss, red eye, eye pain, flashes of light, photophobia, problems with age and contact lenses (Fig. 4). In the menu eye examination there are 8 submenus: visual acuity, visual fields, color vision, pupils, extra-ocular muscles, lamp examination, tonometry and ophthalmoscopy. Each of these menus shows the steps to be followed to diagnose diseases. “Eye Emergency Manual”. This is a guide to recognize the signs and symptoms of the most common eye diseases. In the input menu, their developers claim that it is an emergency eye manual for ophthalmic use of doctors, especially in rural and

Fig. 3 Start of the app “Ophthalmology”

174, Page 8 of 10

J Med Syst (2015) 39:174

Fig. 4 Eye diseases Menu

metropolitan hospitals without ophthalmology service. In addition to the input menu, it has two more: “Emergency” and “References”. The emergencies menu (Fig. 5) presents three options: trauma, red eyes and vision loss. The first has different submenus such as ocular trauma, burns, etc., and each of them has history, examination, treatment and monitoring, and includes photos of diseased eyes. Red eyes menu presents a flow chart (Fig. 6) to decide which diagnosis is the correct and then explains each disease

Fig. 6 Flowchart Red Eyes

with photos. The submenu of vision loss presents a structure similar to this.

Discussion and conclusions

Fig. 5 Emergency Menu

In this paper the different DSS applied to ophthalmology proposed so far are analyzed. In this analysis it is concluded that almost all existing applications focus on the posterior pole of the eye and, from these, most are intended for the diagnosis of diabetic retinopathy. In addition, these applications mostly require image processing or complex ophthalmoscopes, making them inaccessible to places with little material equipment. In terms of market applications, general practitioners are considered the main niche. Applications are directed to primary care and rural areas, to help them know the seriousness of the eye injury, and recommend a visit to the specialist if necessary. This would be very useful and would suppose cost

J Med Syst (2015) 39:174

savings. It also can be used for teaching, so that medical students can get the most likely diagnoses and display images related to these diseases when they introduce a number of signs and symptoms of certain diseases. Due to the lack of DSS in the field of ophthalmology, specifically focused on the anterior pole of the eye, it can be thought that a project of this nature can be consolidated in the previously stated market. Adding to that the exponential growth of Internet use, the use of smartphones with mobile applications in our daily lives and the deployment of 4G networks that will provide higher quality of service to mobile communications, it is hoped that, if launched to the market, an app of MDSS in ophthalmology would become a useful tool for doctors and teachers, becoming an application with great future. The literature review showed that most of the systems developed for ophthalmology focus on the posterior pole of the eye, and being more specific in the early diagnosis of diabetic retinopathy. Besides, most of these systems focus on image analysis using complex gadgets for it, which makes them inefficient in non-specialized consultations. Certain evolution of the ideas was also observed: if in the early years of this century the possibility, yet very remote, of such support systems was posed, now they are numerous and researches begin to consider the possibility of combining them with EHRs, cloud storage or data mining to multiply their efficiency and significantly improve the results. Finally, from the review of commercial applications it should be noted that there are very few DSS in the field of ophthalmology, and most of them are eye guides or material for preparing ophthalmology students. For future lines, from this review the authors are developing a MDSS oriented to primary care physicians and medical students, offering them the opportunity to learn in a more dynamic way the major diseases affecting the anterior pole of the eye (including are pterygium, blepharitis, pinguecula, and hiposfagma, among others) and displaying quality photos of the diseased eye. Acknowledgments This research has been partially supported by Ministerio de Economía y Competitividad, Spain. This research has been partially supported by the ICT-248765 EU-FP7 Project. Conflicts of interest The authors declare that they have no conflict of interest.

References 1. Schuh CJ, de Bruin JS, Seeling W (2013) Acceptability and Difficulties of (Fuzzy) Decision Support Systems in Clinical Practice. IFSA World Congress and NAFIPS Annual Meeting (IFSA/NAFIPS), 2013 Joint, June 24–28, pp 257–262

Page 9 of 10, 174 2. Kilsdonk E, Peute LW, Riezebos RJ, Kremer LC, Jaspers M (2013) From an expert-driven paper guideline to a user-centred decision support system: A usability comparison study. Artif Intell Med;59 (1):5–13 3. Kahai P, Namuduri KR, Thompson H (2006) A Decision Support Framework for Automated Screening of Diabetic Retinopathy. International Journal of Biomedical Imaging 2006:1–8 4. Bourouis A, Feham M, Hossain MA, Zhang L (2014) An intelligent mobile based decision support system for retinal disease diagnosis. Decis Support Syst ;59:341–350 5. Jahns RG, Houck P (2014) The State of the Art of mHealth App Publishing. mHealth App Developer Economics 2014. http:// mhealtheconomics.com/mhealth-developer-economics-report. Accessed 10 November 2014 6. Martínez-Pérez B, de la Torre-Díez I, López-Coronado M,Sainz de Abajo B, Robles M, García-Gómez JM (2014) Mobile Clinical Decision Support Systems and Applications: A Literature and Commercial Review. J Med Syst ;38 (1):4 7. U.S. Department of Health and Human Services, Food and Drug Administration. 2013. Mobile Medical Applications: Guidance for Industry and Food and Drug Administration Staff, 2013 8. IEEE Xplore (2014) IEEE Xplore Digital Library. http://ieeexplore. ieee.org. Accessed 10 November 2014 9. Web of Science (2014) Web of Science, Thomson Reuters. http:// apps.webofknowledge.com/UA_GeneralSearch_input.do?product = UA&search_mode = GeneralSearch&SID = T24y9u4iYdaNoR6csUo&preferencesSaved = Accessed 10 November 2014 10. Scopus (2014) Elsevier Scopus. http://www.scopus.com. Accessed 10 November 2014 11. PubMed (2014) U.S. National Library of Medicine. http://www.ncbi. nlm.nih.gov/pubmed. Accessed 10 November 2014 12. Google Play (2014) Google Play apps. https://play.google.com. Accessed 10 November 2014 13. APP Store (2014) Official Apple Store. http://store.apple.com. Accessed 10 November 2014 14. Prasanna P, Jain S, Bhagatt N, Madabhushi A (2013) Decision Support System for Detection of Diabetic Retinopathy Using Smartphones. 7th International Conference on Pervasive Computing Technologies for Healthcare (PervasiveHealth) 2013, May 5–8, pp 176–179 15. Fraz MM, Remagnio P, Hoppe A, Barman SA (2013) Retinal image analysis aimed at extraction of vascular structure using linear discriminant classifier. International Conference on Computer Medical Applications (ICCMA) 2013, Jan 20–22, pp 1–6 16. Mookiah MRK, Rajendra U, Kuang C, Min C, Ng EYK, Laude A (2013) Computer-aided diagnosis of diabetic retinopathy: A review. Comput Biol Med ;43 (12):2136–2155 17. Noronha K, Acharya UR, Nayak KP, Kamath S, Bhandary SV (2012) Decision support system for diabetic retinopathy using discrete wavelet transform. Proc Inst Mech Eng H ;227 (3):251–261. 18. Bursell SE, Brazionis L, Jenkins A (2012) Telemedicine and ocular health in diabetes mellitus. Proc Inst Mech Eng H 2013;227 (3):251– 261 19. Kumar SJ, Madheswaran M (2012) An improved medical decision support system to identify the diabetic retinopathy using fundus images. J Med Syst ;36 (6):3573–3581 20. Xiao D, Vignarajan J, Lock J, Frost S, Tay-Kearney ML, Kanagasingam Y (2012) Retinal image registration and comparison for clinical decision support. Australas Med J ;5 (9):507–512 21. Skevofilakas M, Zarkogianni K, Karamanos BG, Nikita KS (2010) A hybrid Decision Support System for the Risk Assessment of retinopathy development as a long term complication of Type 1 Diabetes Mellitus. Conf Proc IEEE Eng Med Biol Soc 2010;2010:6713–6716 22. Ortíz D, Cubides M, Suárez A, Zequera M, Quiroga J, Gómez J, Arroyo N (2010) Support system for the preventive diagnosis of

174, Page 10 of 10

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

Hypertensive Retinopathy. Conf Proc IEEE Eng Med Biol Soc 2010;2010:5649–5652 Jegelevicius D, Krisciukaitis A, Lukosevicius A, Marozas V, Paunksnis A, Barzdziukas V, Patasius M, Buteikiene D, Vainoras A, Gargasas L (2009) Network Based Clinical Decision Support System. 9th International Conference on Information Technology and Applications in Biomedicine 2009, Nov. 4–7, pp 1–4 Chia-Ling T, Madore B, Leotta MJ, Sofka M, Gehua Y, Majerovics A, Tanenbaum HL, Stewart CV, Roysam B (2008) Automated Retinal Image Analysis Over the Internet. IEEE Trans Inf Technol Biomed ;12 (4):480–487 Marsolo K, Twa M, Bullimore MA, Parthasarathy S (2007) Spatial Modeling and Classification of Corneal Shape. IEEE Transactions on Information Technology in Biomedicine ;11 (2):203–212 Paunksnis A, Barzdziukas V, Jegelevicius D, Kurapkiene S, Dzemyda G (2006) The use of information technologies for diagnosis in ophthalmology. J Telemed Telecare ;12 Suppl 1:37–40 Roque AC, Mantovani J, Torres I, Galina AC, de Lima PR, Novoa C, Schor P (2004) Lepo: Sistema de Apoio à Decisão Médica em. http:// www.sbis.org.br/cbis9/arquivos/851.pdf. Accessed 10 November 2014 Chiang MF, Boland MV, Margolis JW, Lum F, Abramoff MD, Hildebrand PL (2008) Adoption and Perceptions of Electronic Health Record Systems by Ophthalmologists: An American Academy of Ophthalmology Survey. Ophthalmology ;115 (9): 1591–1597 Acharya UR, Kannathal N, Ng EY, Min LC, Suri JS (2006) A Computer-based classification of eye diseases. Conf Proc IEEE Eng Med Biol Soc ;1:6121–6124 Odufuwa T, Bola O, Solebo L, Low S (2006) Diagnostic decision support in Ophthalmology. J Telemed Telecare 2007;13 suppl 1:44– 46 Xiao-Peng H, Dempere L, Yang GZ (2003) Hot spot detection based on feature space representation of visual search in medical imaging. 4th International IEEE EMBS Special Topic Conference on Information Technology Applications in Biomedicine 2003, April 24–26, pp 261–64 Papageorgiou EI, Huszka C, De Roo J, Douali N, Jaulent MC, Colaert D (2013) Application of probabilistic and fuzzy cognitive approaches in semantic web framework for medical decision support. Comput Methods Programs Biomed ;112 (3):580–598

J Med Syst (2015) 39:174 33. Ping W, Kimb S, Kimb KY, Yangc HJ (2011) Systematic causal knowledge acquisition using FCM Constructor for product design decision support. Expert Systems with Applications ;38 (12):15316–1533 34. Tabachneck-Schijf HJM, Geenen PL (2009) Preventing knowledge transfer errors: Probabilistic decision support systems through the users’ eyes. International Journal of Approximate Reasoning 2008;50 (3):461–471 35. Makam AN, Lanham HJ, Batchelor K, Moran B, Howell-Stampley T, Kirk L, Cherukuri M, Samal L, Santini N, Leykum LK, Halm EA (2014) The good, the bad and the early adopters: providers’ attitudes about a common, commercial EHR. J Eval Clin Pract ;20 (1):36–42 36. Makam AN, Lanham HJ, Batchelor K, Moran B, Howell-Stampley T, Kirk L, Cherukuri M, Samal L, Santini N, Leykum LK, Halm EA (2013) A pilot study of distributed knowledge management and clinical decision support in the cloud. Artif Intell Med ;59 (1):45–53 37. Batra S, Parashar HJ, Sachdeva S, Mehndiratta P (2013) Applying Data Mining Techniques to Standardized Electronic Health Records for Decision Support. Sixth International Conference on Contemporary Computing (IC3) 2013, Aug. 8–10, pp 510–515 38. Kadam T, Chitre V (2012) Three-level HAC on food borne disease and related treatment to help medical DSS. 1st International Conference on Recent Advances in Information Technology (RAIT) 2012, March 15–17, pp 672–676 39. Ahmed S, Abdullah A (2011) E-Healthcare and Data Management Services in a Cloud. High Capacity Optical Networks and Enabling Technologies (HONET) 2011, Dec. 19–21, pp 248–252 40. InSook C, JeongAh K, JiHyun K, Hyun YK, Yoon K (2010) Design and implementation of a standards-based interoperable clinical decision support architecture in the context of the Korean EHR. Int J Med Inform ;79 (9):611–612 41. Madhukumar S, Vijayalakshmi R (2013) Visual dictionary: A decision support tool for DR pathology detection on POI. International Conference on Information Communication and Embedded Systems (ICICES) 2013, Fab 21–22, pp 496–501 42. Rahaman S (2012) Diabetes diagnosis decision support system based on symptoms, signs and risk factor using special computational algorithm by rule base. 15th International Conference on Computer and Information Technology (ICCIT) 2012, Nov. 22–24, pp 65–71 43. Rodriguez Loya S, Kawamoto K, Chatwin C, Huser V (2014). Service Oriented Architecture for Clinical Decision Support: A Systematic Review and Future Directions. J Med Syst ;38:140