Gahangir Hossain

Design, development and performance evaluation of reconfigured mobile Android phone for people who are blind or visually impaired

This paper presents the design, development and performance evaluation of a Reconfigured Mobile Android Phone (R-MAP) designed and implemented to facilitate day-to-day activities for people who are blind or visually impaired. Some of these activities include but are not limited to: reading envelopes, letters, medicine bottles, food containers in refrigerators; as well as, following a route plan, shopping and browsing, walking straight and avoiding collisions, crossing traffic intersections, finding references in an open space, etc. The key objectives were to develop solutions that are light weight, low cost, un-tethered and have an intuitive and easy to use interface that can be reconfigured to perform a large number of tasks. The Android architecture was used to integrate the cell phone camera, image capturing and analysis routines, on-device implementation of robust and efficient optical character recognition (OCR) engine and text to speech (TTS) engine to develop the proposed application in real-time. Empirical analysis under various environments (such as indoor, outdoor, complex background, different surfaces, and different orientations) and usability studies were performed to illustrate the efficacy of the R-MAP. Improved feedback and new functions were added based on usability study results.

Understanding Effects of Cognitive Load from Pupillary Responses Using Hilbert Analytic Phase

Task-evoked pupillary responses reveal the relationship between working memory capacity and its effect on cognitive states. Understanding the effects of cognitive load requires robust analysis of pupillary responses. In this paper, we introduced a Hilbert transform analytic phase based method to compute temporal patterns from pupillary responses. Analysis reveals that sharp change in incorrect task response may be attributed to the cognitive overload. It was also observed that a sharp change and continuation of the ramp in Hilbert unwrapped phase relates to the cognitive dissonance.

Cognitive Load Based Adaptive Assistive Technology Design for Reconfigured Mobile Android Phone

In assistive technology design, it is indispensible to consider the sensory, physical and cognitive level of target users. Cognitive load is an important indicator of cognitive feedback during interaction and became the critical research issue in designing assistive user interfaces, incorporated with smartphone based assistive technology like in the android platform. In this paper, we proposed a cognitive load based user interface integrated with reconfigured mobile android phone (R-MAP) based on user’s cognitive load level. We performed some cognitive tasks within a small group of sighted but blindfolded people and blind people or visually impaired using R-MAP. Based on task performance and cognitive load levels we manually annotated some data of 24 participants and finally applied some machine learning algorithms to automate the mobile interface. Based on our novel design and experimental finding, we recommended that “cognitive load enabled feedbacks based assistive user interface” would be a useful assistive tool for the people who use mobile phone for their daily operations.

Cognitive Ability-Demand Gap Analysis With Latent Response Models

A better understanding of human cognitive ability-demand gap (ADG) is critical in designing assistive technology solution that is accurate and adaptive over a wide range of human-agent interaction. The main goal is to design systems that can adapt with the users abilities and needs over a range of cognitive tasks. It will also enable the system to provide feedback consistent with the situation. However, the latent structure and relationship between human ability to respond to cognitive task (demand on human by the agent) remains unknown. Robust modeling of cognitive ADG will be a paradigm shift from the current trends in assistive technology design. The key idea is to estimate the gap, based on human-agent cognitive task interaction. In particular, latent response model was adopted to quantify the gap. First, we used one parameter Rasch model and extended Rasch model (rating scale model, partial credit model) with dichotomous and polytomous responses, respectively. Residues between expected and observed ability scores were considered as gap parameter in case of dichotomous response. In extended Rasch modeling, response latitudes are considered as an indicator of the gap. Additionally, we performed model fitting, standard error measurement, kernel density estimation, and differential item functioning to test the suitability of Rasch model. Empirical analyses on a number of data set show that proposed analytical method can model the cognitive ADG from dichotomous and polytomous responses. In dichotomous case, the model better fits for mixed responses (combination of easy, medium, and hard) data set rather than monotonic (e.g., only easy) data. Results show that Rasch model can be reliably used to estimate cognitive gap with different cognitive task types.

An audible Bangla text-entry method in Mobile phones with intelligent keypad

Communication through mobile device is the most effective and fastest way that becomes a part of our daily life nowadays. Particularly, Short Message Service (SMS), is one of the most popular and relatively cheaper application of cell phone. In Bangladesh, there has been an enormous growth of mobile users and in this context, it is quite reasonable that people like to send SMS with their own language. Taking these challenges, a few initiatives were taken to introduce Bangla text in Mobile phones, more specifically, in messaging. But unfortunately, Bangla SMS is not popular yet in our country due to the presence of large number of characters and complex script. In fact, it is very much difficult to map Bangla characters efficiently on a mobile with 12 keys only. Thats why, it motivated us to design an intelligent mobile keypad layout, which significantly reduces the number of keystrokes than existing methods, and remove difficulties of entering text. We eliminate traditional multi-tapping on a single button and introduce two key press technique, by arranging characters in two-dimensional matrix. Using this keypad the system becomes faster, reliable and flexible. Predictive text input method is also added in the system for further speeding up messaging. Moreover, our text-entry system ensures accessibility to visual impaired people by avoiding disambiguation caused by multi-tapping and introducing audio feedback for inserting each Bangla characters, which is not supported by existing systems. Finally, after analyzing the result on different sample SMS data, we showed that, our proposed keypad with the predictive input system reduces the number of key presses than the common Bangla keypad by 60.34%.

Spatial Directionality Found in Frontal-Parietal Attentional Networks

Research in last few years on neurophysiology focused on several areas across the cortex during cognitive processing to determine the dominant direction of electrical activity. However, information about the frequency and direction of episodic synchronization related to higher cognitive functions remain unclear. Our aim was to determine whether neural oscillations carry perceptual information as spatial patterns across the cortex, which could be found in the scalp EEG of human subjects while being engaged in visual sensory stimulation. Magnitude squared coherence of neural activity during task states that “finger movement with Eyes Open (EO) or Eyes Wandering (EW)” among all electrode combinations has the smallest standard deviation and variations. Additionally, the highest coherence among the electrode pairs occurred between alpha (8-12 Hz) and beta (12-16 Hz) ranges. Our results indicate that alpha rhythms seem to be regulated during activities when an individual is focused on a given task. Beta activity, which has also been implicated in cognitive processing to neural oscillations, is seen in our work as a manner to integrate external stimuli to higher cognitive activation. We have found spatial network organization which served to classify the EEG epochs in time with respect to the stimuli class. Our findings suggest that cortical neural signaling utilizes alpha-beta phase coupling during cognitive processing states, where beta activity has been implicated in shifting cognitive states. Significance. Our approach has found frontoparietal attentional mechanisms in shifting brain states which could provide new insights into understanding the global cerebral dynamics of intentional activity and reflect how the brain allocates resources during tasking and cognitive processing states.

Rethinking self-reported measure in subjective evaluation of assistive technology

Self-reporting is used as a subjective measure of usability study of technology solutions. In assistive technology research, more than often the ‘a coordinator’ directly assist the ‘subject’ in the scoring process. This makes the rating process slower and also introduces bias, such as, ‘Forer effect’ and/or ‘Hawthorne’ effect. To address these issues we propose to use technology mediated interaction between the ‘subject’ and ‘the coordinator’ in evaluating assistive technology solutions. The goal is to combine both the qualitative and quantitative scores to create a relatively unbiased rating system. Empirical studies were performed on two different datasets in order to illustrate the utility of the proposed approach. It was observed that, the proposed hybrid rating is relatively unbiased for usability study.

Towards a sustainable cyber-physical energy system design

The concept of energy sustainability has been recently introduced in complex cyber-physical systems innovation studies. Previous empirical studies have only focused on the coordination of systems in a large system, but less research has been done on energy efficiency and management towards sustainability. We argue that in terms of energy consumption, a system comprising variety of technologies and applications should be energy sustainable. Hence, we focus on a framework of energy demand in a complex-robotic environment. As a preliminary study, we propose a system dynamics modeling of the energy sustainable complex-structure. As the work is ongoing, some initial simulation results are shown here. Based on our initial studies, we evaluate a number of implications of the management of energy in large-scale cyber-physical systems.

Pattern of Success Vs. Pattern of Failure: Adaptive Authentication Through Kolmogorov–Smirnov (K-S) Statistics

Smartphones have become a basic necessity in lives of all human beings. Apart from the core functionality of communication, these become a medium for storage of sensitive personal information, financial data and official documents. Hence, there is an inevitable need to emphasize on securing access to such devices considering the nature of data being stored. In addition, accessibility and authentication methods need to secure, robust, and user-friendly. This paper discusses an adaptive authentication mechanism with a nonparametric classification approach, Kolmogorov–Smirnov (K-S) statistic, which is coupled with the use of lock pattern dynamics as a secure and user-friendly two-factor authentication method. The data used for experimental exploration were collected from a systematically programmed Android device to capture the temporal parameters when individuals drew lock patterns on the touch screen. Each user has his individualistic way of drawing the pattern, which is used as the key for identifying imposters from valid users.

Study of Phase Relationships in ECoG Signals Using Hilbert-Huang Transforms

This study investigates phase relationships between electrocorticogram (ECoG) signals through Hilbert-Huang Transform (HHT), combined with Empirical Mode Decomposition (EMD). We perform spatial and temporal filtering of the raw signals, followed by tuning the EMD parameters. It can be seen that carefully tuning of EMD filter, it is possible to capture distinct features of non-stationary data. This makes EMD, combined with HHT a valuable tool of complex brain signal analysis and modeling.