voyeurism

welcoming a visiting postgraduate to the team :-)

we are very happy to welcome a visiting postgraduate from East China Normal University as the newest member of our team :-)

our student will be with us over the course of six months, under the auspices of the non-graduating student (research) scheme of our institute.

during her time with us, she will undertake activities themed on Exploring the use of Artificial Intelligence and Data Science for student-centred learning.

we look forward to a mutual time of learning from each other :-)

sharing our students' work with 'A'-level Biology teachers

this morning, our team was very fortunate to be afforded an opportunity to share our students' work on Data Science, AI, and neuroergonomics to 'A'-level Biology teachers.

the sharing was facilitated under the auspices of the in-service course INS2234 Use of microbus and sensors in A-level Biology practical work, and was held at the Science Education Laboratory II of the National Institute of Education (NIE).

our team is very grateful to the Natural Sciences & Science Education Academic Group of the NIE, for extending their trust and hospitality.

:-)

UNESCO / SEAMEO Round Table on Generative AI and Education in Asia-Pacific: the report is published!

i am thrilled to share that my friend and brother Alex Wang has just had a volume he has been working tirelessly on - namely, the report on the UNESCO / SEAMEO Round Table on Generative AI and Education in Asia-Pacific Empowering Minds: A Round Table on Generative AI and Education in Asia-Pacific - has just been published.

the roundtable took place in Bangkok in November.

well done, Alex!

Translating the use of inquiry-based approaches to investigating real-world phenomena using the Internet of Things and Data Science

through God’s grace alone, my team and i received the news today that we have been awarded a grant by the National Institute of Education. titled Translating the use of inquiry-based approaches to investigating real-world phenomena using the Internet of Things and Data Science, the quantum of the grant (inclusive of overheads) is just over US$319000.

the grant is positioned as a translation and scale-up of our earlier grant - Designing for greater authenticity in Geographical Investigations (GI) through local microclimate studies with the Internet of Things (IoT) and open-source environmental sensors.

it is also our second grant themed on Data Science and AI, after Helping students learn coding with automatic feedback and analytics which we were awarded in March 2022.

it looks to be a wonderfully exciting and meaningful journey ahead of us. all thanks be to God.

The Life2Well Project: introductions to the Professional Learning Community at Deyi Secondary School

we were very happy to have been accorded the opportunity today to share our work on Learning at the intersection of AI, physiology, EEG, our environment and well-being (the Life2 Well Project) - particularly with regards its intersections with Generative AI - with the teachers who comprise the Professional Learning Community at Deyi Secondary School.

we would like to express our gratitude to the Senior Leadership team at Deyi for the trust extended to us, and the warm hospitality we received.

thank you!

An exploratory study integrating Deep Learning in digital Clock Drawing Tests on consumer platforms for enhanced detection of mild cognitive impairment

we are very excited to share that our paper on An exploratory study integrating Deep Learning in digital Clock Drawing Tests on consumer platforms for enhanced detection of mild cognitive impairment will be shared at the 26th International Conference on Human-Computer Interaction (HCI).

this year, the conference will be held in Washington DC from the 29th of June till the 4th of July. my co-authors are Bryan Z W Kuok and Malcolm H S Koh.

the abstract reads:

Dementia is a medical condition characterised by cognitive functional decline that affects daily life and social activities (National Institute On Aging, 2022). It is currently the seventh leading cause of death worldwide.

This paper describes an independent research project conceptualised, designed and enacted by a pair of high school students under the mentorship of a senior research scientist at the National Institute of Education in Singapore, from April 2023 to March 2024. The paper documents the development of a prototype of an instrument for the early detection of mild cognitive impairment.

Early detection and diagnosis of cognitive decline is critical to the development and deployment of novel therapeutic intervention for patients with dementia due to Alzheimer’s disease. Clinical detection of dementia currently relies largely on reports from patients, family members or based on cognitive reports from screening tests such as the Mini-Mental State Examination (Folstein, Folstein, & McHugh (1975)) or the Montreal Cognitive Assessment (Nasreddine, et al. (2005)).

These tests however have shown to be relatively insensitive to milder impairments, thus requiring conventional hand-scoring and administration by trained professionals, which brings about subjectivity in the scoring. CANTAB Mobile (Barnett, et al. (2016)) and Cogstate Brief Battery (Maruff, et al. (2013)) are computerised tests that have shown to tackle the problems through automation of the administration and scoring processes. Nonetheless, the process is unnatural and distant from daily life, and combined with lengthy screening sessions, may limit the use of technology, especially for older or uneducated individuals. There exists more invasive and costly methods such as the Cerebrospinal Fluid and Positron Emission Tomography (PET), which provides suboptimal accuracy in prediction of cognitive impairment and limited information.

The Clock Drawing Test (CDT) is a simple pen and paper test that has been used for over 50 years due to its ease of administration (Heerema, 2022). It is non-invasive and inexpensive, yet providing valuable clinical information and diagnostic utility (Shulman, 2000). However, due to its nature of physical administration, it brings about issues such as that of consistency and subjectivity. Digitising the tool would assuage this concern. Indeed, in 2005, Penney and Davis developed a digital clock drawing test, which was subsequently branded as DCTclock (Linus Health, 2022).

While Penney’s & Davis’s (2005) DCTclock offered a significant improvement of detection of early cognitive impairment (Souillard-Mandar, et al. (2021)), development of such proprietary technology tends to be expensive. It was from this framing that the authors explored the prototyping of a digital Clock Drawing Test (dCDT) application on platforms appropriate for non-specialist devices (such as iOS and Android) in order to enable timely information updates and subsequent analysis of data.

Thus far, the application integrates various functions, including patient information input, the Clock Drawing Test, image classification using machine learning models, result output, and data logging. The aim is to simplify the screening process for both patients and medical professionals, thereby aligning with the Human-Computer Interaction (HCI) principle of simplicity.

To extend the affordance structures of the application, the authors simulated drawing on pen and paper by utilising Apple Pencil as the digital stylus and iPad as the drawing canvas. As is well known, the stylus and the tablet are equipped with sensors capable of collecting a sufficient amount of information from which to glean meaningful conclusions regarding patients’ drawings. Additionally, the authors established a database using Realm Studio, which is a visual tool enabling medical professionals to intuitively view, edit and model patients’ data.

To maintain design consistency, the authors studied the Human Interface Guidelines (Apple (2023)). This approach ensures that the application aligns with best practices across the industry, making it user-friendly and familiar to users while differentiating elements that serve specific functions, such as text boxes and action buttons.

At the time of writing, the authors are actively exploring advanced machine learning models to analyse additional parameters, improving the evaluation of cognitive impairment.

The significance of this work lies in addressing the general lack of awareness of dementia due to a dearth of education (Sharp & Gatz (2011)). In addition, social perception arising from negative depiction of dementia in popular culture and cultural stigma (Low & Purwaningrum (2020)), and lower socioeconomic status (Petersen et al. (2021)) have also been identified as risk factors for dementia.

A study published in The Lancet Public Health reported that the number of adults aged forty years and older living with dementia worldwide is expected to nearly triple from an estimated 57 million in 2019 to 153 million in 2050 given the projected trend in population ageing and population growth (Nichols et al. (2021)). Therefore, healthcare systems must prepare to provide timely and accurate diagnoses and have the resources in place for efficient care. This paper describes one such possible aspect through which this may be achieved.

access our portfolio with GPinTuitions - now live in the GPT Store!

further to our post from two months' ago, we are very happy to share that you our custom GPT in to our portfolio - GPinTuitions - is now accessible through the GPT Store, which just launched today!

[30 May: GPinTuitions is now available to all ChatGPT users]

sharing with visiting teachers as part of the International Science Youth Forum (ISYF)

my team and I were very happy and excited today to have been given the privilege to share the work of our students to a group of teachers as part of the International Science Youth Forum (ISYF) 2024.

the teachers hailed from local schools within Singapore, as well as from schools beyond our borders.

we shared the following student-initiated projects:

- Exploring local ornithological identification with Deep Learning (Edwin Ong and Liow Shang Yu);
- Designing an Autonomous Platform using Hobbyist Robotics Embedded Computing for Performing Geospatial 3D Mapping (Ong Yi Xin and Christopher Cheng);
- Exploring the Potential Affordances of Augmented Reality (AR) Game-Based Learning in Campus Heritage Preservation (Vu Pham Bao Anh and Ying Qinyi);
- Investigating the relationship between microclimate factors and human exercise performance (Dinh Nho An and Quek Yi Ming); and
- Investigating the effect of microclimate on human's physiological health using Electrodermal activity (Nguyen Duc Minh Anh and Nguyen Thien Minh Tuan).