iPrevention: Towards a Novel Real-time Smartphone-based Fall Prevention System AKM Jahangir Alam Majumder 1 , Farzana Rahman 1 , Ishmat Zerin 1 , Ebel Jr. William 2 , Sheikh Iqbal Ahamed 1 1 Department of MSCS, Marquette University, Milwaukee, WI 53233, USA 2 Department of ECE, Marquette University, Milwaukee, WI 53233, USA {akmjahangir.majumder, farzana.rahman, ishmat.zerin, william.ebeljr, sheikh.ahamed}@mu.edu ABSTRACT Falling remains one of the leading causes of hospitalization and death for the elderly all around the world. The considerable risk of falls and the substantial increase of the elderly population have stimulated scientific research on smartphone-based fall detection systems recently. Even though these systems are helpful for fall detection, the best way to reduce the number of falls and their consequences is to prevent them from happening in the first place. Therefore, our focus is on fall prevention rather than fall detection. To address the issue of fall prevention, in this paper, we propose a smartphone-based fall prevention system that can alert the user about their abnormal walking pattern. Most current systems merely detect a fall whereas our approach attempts to identify high-risk gait patterns and alert the user to save them from an imminent fall. Our system uses a gait analysis approach that couples cycle detection with feature extraction to detect gait abnormality. We validated our approach using a decision tree with 10-fold cross validation and found 99.8% accuracy in gait abnormality detection. To the best of our knowledge, we are the first to use the built-in accelerometer and gyroscope of the smartphone to identify abnormal gaits in users for fall prevention. Categories and Subject Descriptors H.4 [Information Systems Applications]: Miscellaneous General Terms Experimentation, Design, Human Factors. Keywords Fall Prevention, Gait, Smartphone, Accelerometer, Gyroscope, Motion sensor. 1. INTRODUCTION Falls in the elderly are a very common occurrence that can have dramatic health consequences. For people over 70-75 years old, the estimated incidence of falls is over 30 percent per year [1]. Nearly half of nursing home patients fall each year, with 40 percent falling more than once [2]. However, it is the impact rather than the occurrence of falls in older adults, which is of most concern. Older adults are typically frailer, more unsteady, suffer from various diseases, have abnormal and unsteady walking patterns due to aging, and thus are more likely to be injured than toddlers and athletes, who also fall regularly. Falls can cause physical injury to people, especially to the elderly, including fractures, head injuries, or serious lacerations. Falls not only cause physical injuries, but also have dramatic psychological consequences that reduce elderly peoples’ independence [3, 4]. This can lead to an avoidance of activity that can bring about a pattern of increasing isolation and health deterioration. In addition to the psychological and physical trauma of a fall, incidents can involve various other issues as well: the cost of hospitalization, and physical therapy. The considerable risk of falls and the substantial increase of the elderly population have recently stimulated scientific research on fall detection. Although fall detection systems cannot directly prevent falls, detection can help reduce the risk of fallen patients who have been left immobilized or unconscious by a fall from being left untreated for an extended period. Though there has been a lot of research on automatic fall detection, the area of fall prevention has been understudied. Fall prevention is very challenging since to prevent a fall, first we need to identify the patterns that can lead to a fall. Current work on automatic fall detection methods can be classified into three main categories in terms of the sensors they use: video- based methods [6], acoustic-based methods [7] and wearable sensor-based methods [5]. A majority of fall detection systems require some specific hardware or software design [8, 9]. This increases cost and limits its usage to the wealthiest people of society only. Many systems also have significant installation and training times, which causes the average person to tend to reject the system. With the recent improvements in mobile technology, the costs of smartphones have decreased but their computational capabilities have increased. As self-contained devices, smartphones present a mature hardware and software environment for developing various fall detection systems. Smartphone-based fall detection systems can function almost everywhere, since mobile phones are highly portable. Though it is argued that elderly people may not accept a mobile phone based system, it is also argued that elderly people may prefer to have a single phone with self-contained fall detection functionality rather than to carry a separate fall detection device on their bodies. Currently, most smartphones now have sensors to observe acceleration, location, orientation, ambient lighting, sound, imagery, etc. [10, 11]. Since integrated sensors along with the mobile platform are ideal for developing an application to detect falls, researchers have already developed some smartphone-based fall detection systems [12, 13]. However, in all of these previous studies, the system can detect a fall only after it has already occurred and the system sends an alarm to the caregivers for immediate help. Even though these fall detection systems are helpful, the best way to reduce the number of falls and their consequences is to prevent them from happening in the first place. We believe that the best way to reduce the number of falls is to alert the users about their abnormal gait/walking and the possibility of falling. The term “gait” refers more specifically to the manner or pattern of walking. Gait and balance disorders are common in older adults and are a major cause of falls in this population. Almost 30 percent of the elderly people of the age 65 or more report difficulty walking or Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. SAC’13, March 18-22, 2013, Coimbra, Portugal. Copyright 2013 ACM 978-1-4503-1656-9/13/03…$10.00. 513