• Digital-Health-Conference-2020


​Agenda of the conference Digital Health 2020

  • Day 1Monday, January 20th
  • Day 2Tuesday, January 21st
  • Day 3Wednesday, January 22nd
8:30 am

Registration ( Building 19, Hall 1)

Building 19, Hall 1 08:30 - 09:00 Details

9:00 am

Coffee & Breakfast (Building 19, Hall 1)

Building 19, Hall 1 09:00 - 09:30 Details

9:30 am

Welcome Address

  • Prof. Takashi Gojobori, KAUST

    Prof. Takashi Gojobori

Building 19, Hall 1 09:30 - 09:35 Details

Prof. Takashi Gojobori, KAUST
9:35 am

Opening Remarks

Building 19, Hall 1 09:35 - 09:50 Details

Elmootazbellah Elnozahy, KAUST
Session 1 : Digital Health and vision 2030 (Chair Prof. Takashi Gojobori)
9:50 am

KEYNOTE LECTURE: The KAUST Smart Health Initiative from “bench to bedside”: Promoting research collaborations in the Kingdom

With a collective mission at King Abdullah University of Science and Technology (KAUST) to drive cutting-edge research, innovation, and to create local/global impact through applied sciences, KAUST has launched a new strategic initiative in the field of “Smart Health”.

Through the integration of basic scientists, clinical investigators, and innovative technologies, the KAUST Smart Health Initiative aims at creating the ecosystem required to foster highly collaborative “bench to bedside” research and capacity building programs with the greatest potential to transform the practice of healthcare, improve human health, and drive both our fundamental understanding of disease mechanisms and precision medicine in the Kingdom.

The Initiative has begun on January 1, 2020, with a pre-launch phase focused on identifying clinical partners and projects that will form the building blocks for the larger-scale program.

  • Prof. Pierre J. Magistretti, KAUST

    Prof. Pierre J. Magistretti

Building 19, Hall 1 09:50 - 10:25 Details

Prof. Pierre J. Magistretti, KAUST
10:25 am

CHAIRMAN'S SPECIAL GUEST SPEAKER: How can genetics, genomics, and related technologies create value and increase efficiency of R&D?

The rise of genomics technologies over the past two decades has enabled the high-throughput interrogation of the entire human genetic complement. We can also now interrogate the whole human exome or genome , which reveals which genes are actively being converted into their protein products at a particular time or under specific conditions (such as healthy vs. diseased states). Such technologies have also facilitated the advent of a whole new field of human

  • Dr. Malak Abedalthagafi , KACST-KFMC

    Dr. Malak Abedalthagafi

Building 19, Hall 1 10:25 - 10:55 Details

Dr. Malak Abedalthagafi , KACST-KFMC
10:55 am

Coffee Break

Building 19, Hall 1 10:55 - 11:10 Details

11:10 am

Innovation, consumer health informatics and future of health care

Innovation, consumer health informatics and future of health care

Building 19, Hall 1 11:10 - 11:35 Details

Prof. Ahmed Albarrak , College of Medicine, King Saud University
11:35 am

Healthcare Informatics: Towards the evolution of Personalized Care to Population Health Management

This session will touch base on the evolving role of healthcare informatics and how it will transform and reshape healthcare from current curative perspective towards prevention and managing large scale populations, with notion to the current transformative care delivery in our local context, to ensure seamless health and care starting from a personalized care extending to population health management.

  • Mohammed Alhefzi , King Faisal Medical City for Southern Regions

    Mohammed Alhefzi

Building 19, Hall 1 11:35 - 12:00 Details

Mohammed Alhefzi , King Faisal Medical City for Southern Regions
12:00 pm

Day 1 Lunch Break

Lunch will be served at Campus diner

Campus Diner 12:00 - 13:15 Details

Session 2 : Digital Health and Wellness (Chair Prof. Xin Gao)
1:15 pm

KEYNOTE LECTURE: Applied Digital Health research

The increased reliance of health systems on the digital record as the primary mechanism for storing data on consultations and other health interactions has opened new opportunities for research, healthcare innovation, and health policy. The electronic health record (eHR) is now ubiquitous in many countries, in hospital and primary care settings, and in some countries their health systems in terms of reporting patient care activity are essentially ‘paperless’.

Health systems globally are also facing accelerating challenges as they seek to deliver better value healthcare against the background of increasing levels of chronic disease, ageing populations, financial pressures and demands on public spending. Digital health tools and services are held up to be part of the solution to these challenges, potentially offering low-cost and patient-centred solutions.

There has been huge investment in Big Data research in health, particularly in relation to digitised imaging and automated reporting and predictive modelling using phenotypic and increasingly genetic data. There have also been similar gains in more applied research that explores the potential of accessing the huge quantum of data held in the eHR, and linkage of these data to other national or regional databases, such as mortality records or cancer data. This session will explore some of the applications for routine data research, illustrated by projects that have resulted in research success and better healthcare.

This will include the exemplars of using large eHR platforms and prescribing data platforms to create infrastructure for i) common disease surveillance, such as the UK RCGP RSC; ii) generation and validation of disease risk assessment tools, such as QRisk scores; iii) pragmatic electronic follow up trials; iv) within practice systems dashboard feedback reports, eg data normalised to regional and national rates on prescribing and investigation physician activity; v) traditional epidemiological linkage studies; and vi) linkage to long term phenotypic follow up of established disease cohorts.

Building 19, Hall 1 13:15 - 13:50 Details

Prof. Richard Hobbs , University of Oxford
1:50 pm

Closing the Data Loop in Healthcare Through Automation: Toward Safer, Higher Quality, Faster Services

Every day, data are being generated in healthcare by hospitals. In today’s medicine, the majority of patient’s data are generated inside clinical facilities. Providers then utilize data in order to make conclusions and make decisions for patients.

In future medicine, this data cycle will be closed using automation and taking advantage of all the data that patients generate outside clinical facilities as well. Closing the data loop through automation will help bring faster decisions by eliminating the middlemen in healthcare delivery processes. It also can help us focus on what matters to patients by personalizing healthcare based on the data that patients generate.

Automating clinical decisions based on patient’s data will not only help sick patients but also can be utilized to promote health for healthy people. Closing the data loop through automation comes with its challenges and barriers, which we will discuss in this topic.

Building 19, Hall 1 13:50 - 14:15 Details

Dr. Nasser Aljehani , King Fahad Medical City
2:15 pm

Wateen, The National Blood Donation platform in Saudi Arabia

Building 19, Hall 1 14:15 - 14:40 Details

Dr. Ahmad Alonazi, MD , Wateen, Application
2:40 pm

Coffee Break

Building 19, Hall 1 14:40 - 14:55 Details

2:55 pm

Broad and Deep Learning of Big Heterogeneous Health Data for Precision Medicine

Broad and Deep Learning of Big Heterogeneous Health Data for Precision Medicine

Building 19, Hall 1 14:55 - 15:20 Details

Prof. Vincent S. Tseng , National Chiao Tung University, Taiwan
3:20 pm

New Nanopore Sequencing Strategies for Accurate, Accessible and Accelerated Genomic Diagnosis

The Oxford Nanopore sequencing technology is unique in its ability to deliver ultra-long single-molecule reads of DNA and RNA. Nanopore sequencing can be done in real-time on a portable device, which makes it an attractive platform for personalized genome medicine.

The long reads of nanopore sequencing have been exploited in scaffolding genomes, especially in repetitive DNA, and in the detection of structural variations in the genome. However, the low raw-read accuracy of nanopore sequencing hampers its application in genomic diagnosis.

We have developed several new strategies that use targeted locus amplification, single-molecular consensus sequencing, and deep-learning-based data analysis tools to enable rapid and accurate determination of single nucleotide variants and structural variants at subclonal levels. We d these technologies to rare mutation detection and analysis of genome integrity after CRISPR genome editing.

Building 19, Hall 1 15:20 - 15:45 Details

Prof. Mo Li, KAUST
3:45 pm

PANEL DISCUSSION - (Moderator - Takashi Gojobori) (Co-Moderator - Fadwa Attiga)

Building 20 15:45 - 16:45 Details

Prof. Takashi Gojobori, KAUST Dr. Nasser Aljehani , King Fahad Medical City Prof. Robert Hoehndorf , KAUST Prof. Ahmed Albarrak , College of Medicine, King Saud University Dr. Malak Abedalthagafi , KACST-KFMC Dr. Fadwa Attiga, Ph.D., Global Oncology Inc. Dr. Ahmad Alonazi, MD , Wateen, Application
5:00 pm

Student Poster Session and Finger Buffet - Open to All

location Campus Library

Campus Library 17:00 - 18:30 Details

8:30 am

Coffee and Breakfast

Building 19, Hall 1 08:30 - 09:30 Details

Session 3 : Digital Health & Food and Environment (Chair Prof. Xiangliang Zhang)
9:30 am

KEYNOTE LECTURE: From healthy plants to healthy humans: what we can learn from studying beneficial microbes

Our research focuses on how plants can survive under abiotic or biotic stress conditions. For this purpose, we isolated more than 2500 beneficial microbes and currently investigate their potential to help grow crops in arid regions of the world (www.darwin21.org). Our results show that microbes can improve crop yield and help crops to withstand diseases. But beneficial microbes also provide the human body with essential elements to establish a healthy gut microbiome.

By using sophisticated techniques and bioinformatic programs, we try to identify specific functions of different microbial strains with respect to plant and/or human health. Although this research is just at the beginning, the overwhelming results promise a revolution in the food sector and human health care.

Building 19, Hall 1 09:30 - 10:05 Details

Prof. Heribert Hirt , KAUST
10:05 am

Leveraging digital platforms to improve healthy diets and increase environmental awareness

The NEOM team is building a city of the future, focused on integrated the latest technological advances in digital and biotech platforms across all sectors. Today, the Food sector is being challenged and changed drastically as technology advances from digital and medical sciences improve crop productivity, decrease reliance on labor and bring about environmental awareness.

Digital technologies will be the heart of the Future of Food. Two major ways in which we foresee the impact are through:

1. Sustainable Sourcing platforms built using collections of digital touchpoints throughout the food system journey.
2. Nutritional labeling and nudges based on the desired lifestyle.

Both of these advances are predicated on technologies but have impacts on human and environmental health

  • Dr. Babar Khan , NEOM

    Dr. Babar Khan

Building 19, Hall 1 10:05 - 10:30 Details

Dr. Babar Khan , NEOM
10:30 am

Coffee Break

Building 19, Hall 1 10:30 - 10:45 Details

10:45 am

Machine Learning Reveals Crop Genome Evolution: Better Domestication Leads to Healthier Diet

Domestication is anthropogenic evolution that fulfills mankind’s critical food demand. As such, elucidating the molecular mechanisms behind this process promotes the development of future new crops. Here we show the genome-wide introgressive region map as traces of domestication of Asian rice, by utilizing 4,587 accession genotypes with a stable outgroup species, particularly at the finest resolution through a machine learning-aided method.

The results provide a definitive answer to a long-standing controversy in plant science, and new breeding designs and practices based on the introgressive genomic map with the aim of serving healthy diet.

Building 19, Hall 1 10:45 - 11:10 Details

Dr. Hajime Ohyanagi, KAUST
11:10 am

Alternative green expression systems for application in tailor-made medicines

Microalgae are globally important primary producers that use light energy and CO2 to generate biomass and valuable bio-products. These organisms grow in simple media solutions, requiring only light, CO2, and some trace minerals while requiring no complex additives compared to mammalian expression systems.

Expanding capacities in genetic engineering in algal hosts offers incentives for the development of low-cost bio-processes using these photosynthetic eukaryotes and a wide range of opportunities for tailoring production of medicinal products. We developed a novel transgene design strategy for the microalga Chlamydomonas reinhardtii through synthetic intron incorporation.

This presentation will overview prospects for customized terpenoid and recombinant protein expression targets, and deliver a vision for future development of low-cost algal synthetic biotechnology as it applies to the medical field.

  • Prof. Kyle Lauersen , KAUST

    Prof. Kyle Lauersen

Building 19, Hall 1 11:10 - 11:35 Details

Prof. Kyle Lauersen , KAUST
11:35 am

Group Photo session with Conference Speakers

Follow the Photographer to the Spot for Photography

Follow instructions to location 11:35 - 11:50 Details

11:50 am

Day 2 Lunch Break

Lunch will be served at Campus Diner

Campus Diner 11:50 - 13:15 Details

Session 4 : Digital Health and Biotechnology (Chair Stefan Arold)
1:15 pm

KEYNOTE LECTURE: Molecular pathogenesis of endometriosis and cancer development

High-throughput sequencing technologies revolutionized medical genomic research, which enabled us to proceed “sequence-based medicine”. Endometriosis is a common disease affecting about 8 % of women, also might develop ovarian cancer in some cases.

Retrograde menstruation is well-known as the origin of endometriosis but there is no molecular-based supports. We focused on somatic mutation profiles in both endometriotic and normal uterine endometrial epithelium samples to prove the retrograde menstruation hypothesis leading to the pathogenesis of endometriosis. We analyzed whole-exome and target-gene sequencing data derived from 107 ovarian endometriotic epithelium and 82 normal uterine endometrial epithelium samples.

Although common cancer-associated mutations were detected in both tissues, their distributions of mutant allele frequency (MAF) in endometriotic epithelium were significantly higher than in normal endometrium. Particularly, steep increase in MAF of mutations on KRAS in endometriotic epithelium was observed suggesting that endometrial tissues harboring KRAS mutations were transported in a retrograde direction to the ovarian surface, where the specific KRAS mutations gave them selective advantages at this and other ectopic sites, leading to the development of endometriosis and widespread distribution of the clone across the endometriotic lesions.

In particular case, we conducted exome sequencing for a set of samples including normal uterine endometrium, distant endometriosis, atypical endometriosis, stromal cells in cancer, and ovarian clear cell cancer (OCCC) tissues from a 56-year-old patient. We analyzed the mutant allele frequencies of somatic mutations in cancer-driver genes present in each of the samples, which enabled us the identification of a sequential genomic footprint that is shared in some of the tissues.

Furthermore, we recognized a directional evolution pattern that denotes the importance of the retrograde menstruation theory in the development of endometriosis, and subsequently, its evolution into OCCC.

In conclusion, our genomic studies disentangled the riddle of the origin of endometriosis supporting Sampson’s retrograde hypothesis with a century-old history.

Building 19, Hall 1 13:15 - 13:50 Details

Prof. Ituro Inoue, National Institute of Genetics
1:50 pm

Gene regulation and dysregulation in human diseases

"DNA makes RNA makes protein." To meet the demands of complex organism development and the appropriate response to environmental stimuli, every step in these processes needs to be finely regulated. Dysregulation could result in pathological conditions. Our lab is interested in quantitative understanding of the molecular mechanism underlying different layers of gene regulation and their dysregulation in human diseases including cancer.

In this talk, I will present our recent study of transcriptional and post-transcriptional dysregulation during EMT process using colorectal cancer cell as a model. Here, by integrating a variety of genomics, transcriptomics and epigenomics data, we identified a number of metastasis-related transcription factors that were transcriptionally or post-transcriptionally dysregulated during EMT process.

Their effects could be validated by using in vitro and in vivo assays. Further functional analysis of their target genes revealed novel pathways involved in EMT process. Finally, indicated by patient survival data of various cancer, these transcription factors may serve as markers with high prognostic potential.

Building 19, Hall 1 13:50 - 14:15 Details

Prof. Wei Chen , Southern University of Science and Technology
2:15 pm

Coffee Break

Building 19, Hall 1 14:15 - 14:30 Details

2:30 pm

Computational genomics of brain tumors: glioma biomarker identification and characterization through multi-omics integrative molecular profiling

Glioma, one of the most lethal human malignancies, represents almost 80% of malignant brain tumors and exhibits low resection rate and high recurrence risk. With the rapid advancement of sequencing technologies, there is an increasing number of high-throughput studies on glioma, resulting in massive multi-omics multi-cohort data generated from different projects and different laboratories throughout the world.

Therefore, it has become critically important on how to make full use of these valuable data for comprehensive integrative characterization of glioma biomarkers. In this study, we collected a large-scale assemble of multi-omics multi-cohort datasets from public resources, involving a total of 17,022 samples across 19 independent studies. We established a methodological strategy on integrative identification of biomarkers with higher specificity and feasible detectability from periphery and identified that PRKCG (Protein Kinase C Gamma) features higher specificity in brain and detectability in CSF (Cerebrospinal Fluid).
Through comprehensive molecular characterization of PRKCG based on multi-omics analyses in RNA expression, DNA methylation and copy number variation, we revealed that PRKCG has the significant potential in glioma diagnosis, prognosis and treatment prediction as testified on multiple independent discovery and validation datasets.

Unlike existing biomarkers that were mostly discovered at single omics level and with limited samples, we found that multi-omics molecular profiles of PRKCG are highly associated with glioma across different populations, bearing great potential for glioma diagnosis, prognosis and therapy.

  • Prof. Zhang Zhang

    Prof. Zhang Zhang

Building 19, Hall 1 14:30 - 14:55 Details

Prof. Zhang Zhang
2:55 pm

Digital Health - What Can We Expect From AI in Regenerative Medicine

Digital Health is a conglomerate of several complementary advancements in computational sciences, artificial intelligence (AI), big data analysis, the OMICS fields as well as biomedical technologies with the overall aim to improve the health of individuals and the society’s well-being.

Digital Health might succeed with its promise to completely revolutionize the entire healthcare system. This is particularly important when considering challenges such as shortage of doctors, limited healthcare resources and rising demands from an aging population in the developed world, but similar important in the developing countries due to the lack of developed infrastructure. Analyzing volumes of data and scanning for patterns and relationships AI is an apt technological tool for big data management that could impact and support personalized medicine in areas such as regenerative medicine, prediction of therapeutic compounds by deep learning and AI-assisted 3D bioprinting.

Building 19, Hall 1 14:55 - 15:20 Details

Prof. Charlotte A. E. Hauser, KAUST
7:00 pm

Gala Dinner - Invitation Only

Dinner will be served at Al-Marasa Restaurant

Yacht Club restaurant 19:00 - 20:30 Details

8:30 am

Coffee and Breakfast

Building 19, Hall 1 08:30 - 09:30 Details

Session 5 : AI & Computational Resources (Chair Dr. Katsuhiko Mineta)
9:30 am

KEYNOTE LECTURE: Predictive and preventive medicine by dynamic network biomarker

Considerable evidence suggests that during the progression of complex diseases, the deteriorations are not necessarily smooth but are abrupt, and may cause a critical transition from one state to another at a tipping point. Here, we develop a model-free method to detect early-warning signals of such critical transitions (un-occurred diseases), even with only a small number of samples.

Specifically, we theoretically derive an index based on a dynamic network biomarker (DNB) that serves as a general early-warning signal indicating an imminent sudden deterioration before the critical transition occurs [1][2]. Based on theoretical analyses, we show that predicting a sudden transition from small samples is achievable provided that there are a large number of measurements for each sample, e.g., high-throughput data.

We employ gene expression data of three diseases to demonstrate the effectiveness of our method for predictive and preventive medicine. The relevance of DNBs with the diseases was also validated by related experimental data (e.g., liver cancer, lung injury, influenza, type-2 diabetes) and functional analysis. DNB can also be used for the analysis of nonlinear biological processes.

Building 19, Hall 1 09:30 - 10:05 Details

Prof. Luonan Chen, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences
10:05 am

Predictive Modeling and Machine Learning – Disruptive Tools enabling Digital Health ? – Challenges and Opportunities

There is a fair amount of hope, if not hype, to the extent that computational techniques can transform our understanding of biology, early discovery of disease, and augmenting health. This includes improved access to healthcare, tools for supporting a healthy lifestyle, and enabling research tools for the clinician as well as for the fundamental life-scientist.

Using a couple of real-world examples from our work – monitoring the health of an athlete, building web-based research interfaces for clinicians, and predictive analytics from hospital records (in-patient-out-patient data) – I’l discuss challenges and opportunities in this space.



Building 19, Hall 1 10:05 - 10:30 Details

10:30 am

Coffee Break

Building 19, Hall 1 10:30 - 10:45 Details

10:45 am

Learning to diagnose Cancer

In this talk, I will discuss the use of natural language processing methods to read scientific literature and learn the implicit features that cause certain genetic mutations to be pathogenic. Precisely, our method ingests the bio-medical literature and produces its fixed representation via exploiting state of the art NLP methods like word2vec and tf-idf.

These representations are then fed to machine learning predictors to identify the pathogenic versus neutral variations. I will also discuss the use of deep learning methods to classify breast cancer into four classes. I will discuss the use of scaled networks to diagnose the various stages of breast cancer. Both of our methodologies significantly outperform previous state-of-the-art studies on publicly available datasets.

Building 19, Hall 1 10:45 - 11:10 Details

Dr. Hammad Naveed , National University of Computer & Emerging Sciences
11:10 am

Logic, Logic, Logic, and Medicine

Knowledge representation is a sub-field of AI which studies
how to represent information about a domain so that is can be utilized for a wide range of tasks. In particular, the life sciences have created a large amount of formalized knowledge bases. In my talk, I will show how to use information in formalized knowledge bases as background knowledge in statistical analyses and machine learning.

I will discuss an algorithm to construct a map from formal theories into vector spaces that preserve the model-theoretic semantics of the theories while enabling new operations within the vector space. Combining methods from knowledge representations with machine learning can be used to generate explanations and exploit background knowledge, which is particularly important in knowledge-intensive disciplines such as biology and medicine.

Building 19, Hall 1 11:10 - 11:35 Details

Prof. Robert Hoehndorf , KAUST
11:35 am

Lunch Break Day 3

Lunch will be served at Campus Diner

Campus Diner 11:35 - 13:00 Details

Session 6 : Spotlight on Young Talent (Chair Prof. Robert Hoehndorf)
1:00 pm

Isolation and characterization of cellulase-producing microorganisms from the Red Sea environment

Cellulolytic microorganisms are considered as a key player in various environments to degrade the plant biomass. Cellulases are used in various applications in industries like biofuels, food and beverages and medical applications.

These microorganisms can be isolated from various environments such as soils, insect gut, mammalian rumen, and oceans. The Red Sea has a unique environment in terms of high seawater temperature, high salinity, low nutrients, and high bio-diversity. However, there is little information regarding cellulases genes in the Red Sea environment.

The aim of the study was to examine if the Red Sea can be a potential resource for bio-prospecting of microbial cellulases, by isolation and characterization of cellulase-producing microorganisms from the Red Sea environment. Three bacterial strains were successfully isolated from plankton fraction and seaweed surface. The isolated strains were identified as Bacillus paralichiniformis and showed strong cellulase activity. These results suggest that these four isolates secreted active cellulases.

Next, we compared the expression of cellulase genes under cellulase-inducing and non-inducing condition, and identified several cellulase genes that were upregulated during the cellulolysis from each isolate. These genes are expected to play important roles in their cellulolysis.

Building 19, Hall 1 13:00 - 13:20 Details

Siham Fatani, KAUST
1:20 pm

Assist antibiotic resistance detection with deep learning

Antibiotic resistance has become one of the most urgent threats to global health, as more drugs are losing sensitivity to bacterium they were designed to kill. When investigating it, researchers usually need to identify and annotate antibiotic resistance genes (ARGs) from environmental or clinical samples.

Despite the existence of several computational tools for performing ARG annotation, most of them rely on sequence alignment, which can result in false negatives and biased prediction because of the incompleteness of the databases. In addition, most existing computational tools provide no information about the mobility of genes and the underlying mechanisms of resistance.

To address such limitations, we propose an end-to-end Hierarchical Multi-task Deep learning framework for Antibiotic Resistance Gene annotation(HMD-ARG), taking raw sequence encoding as input and then annotating ARGs sequences from three aspects: resistant drug type, the underlying mechanism of resistance, and gene mobility. HMD-ARG can potentially serve as a tool for detecting antibiotic resistance with high accuracy.

Building 19, Hall 1 13:20 - 13:40 Details

Yu Li (Ph.D. student) , King Abdullah University of Science and Technology (KAUST)
1:40 pm

Participatory design to successfully develop digital healthcare with multiple stakeholders

Digital health (DH) holds many promises for the future of healthcare both in Saudi Arabia and other countries around the world. However, looking at the implementation of DH, as for example electronic medical records, it remains challenging to achieve the promised goals. Whereas the impact of DH is often measured at patient level, the impact on health care professionals is often overlooked.

Health care professionals using increasingly complex DH systems often find these systems “user unfriendly”, and they are having an increased risk of burnout because of the increasing amount time they have to spent working with them. Considering the increasing interest of spreading acritical intelligence, coupled with big data analytics and genomics, one may assume that this the pressure will continue to rise for health care professionals.

In addition, societal, ethical and financial questions are raised with the implementation of these technologies. How can an increasingly elderly population use digital interfaces of DH, i.e. apps (societal)? Who should govern our most private, genomic data (ethical)? How can these expensive technologies remain affordable (financial)? Considering these various questions about the development of DH, a method has been suggested to incorporate different perspectives from users (including health care professionals and older patients), society, ethics, and finance in the design of DH. Participatory design is the activity of involving designers and non-designers in a democratic way in the design process, whereby mutual learning and creativity are key.

There are various ways to conduct PD and the theory of PD is still in development. There are four phases in PD: exploration of the problem, developing a definition of the problem, creation of a solution and evaluation and testing of the solution. Each step has key tools. The focus of this talk will be on providing an overview of PD methods, tools and theory to develop DH successfully.

This will help researchers and practitioners to include multiple perspectives in the development of DH. Aspects for future research will also be highlighted focusing on the development of PD theory.

Building 19, Hall 1 13:40 - 14:00 Details

Pieter Vandekerckhove (Ph.D. student), Erasmus University of Rotterdam, Erasmus School of Health Policy & Management
2:00 pm

Contributions to Digital Health in Low and High Level

Digital health (DH) is an important topic these days since it provides a lot of help to the smart world that we are approaching. This study contributes to digital health on different levels by introducing methods based on machine learning (ML).

The first contribution concerns developing generic models for the recognition of different genomic signals and regions (GSR) within eukaryotic DNA, which is a low level in DH. We developed DeepGSR, a systematic framework that facilitates generating ML models to predict GSR with high accuracy. To the best of our knowledge, no available generic and automated method exists for such task that could facilitate the studies of newly sequenced organisms and detect some diseases that are caused by specific genes. The prediction module of DeepGSR uses deep learning algorithms to derive highly abstract features that depend mainly on proper data representation and hyperparameters calibration. DeepGSR, which was evaluated on recognition of PAS and translation initiation sites (TIS) in different organisms, yields a more straightforward and more precise representation of the problem under study, compared to some other hand-tailored models, while producing high accuracy prediction results.

Secondly, we focus on deriving a model capable of facilitating the early detection of breast cancer, which is a high level in DH. We provide E-Detect, a full system with frontend (website) and backend (deep learning model), that allows the users to upload the mammograms and collect the results of the probability of having breast cancer. E-Detect uses both image processing and deep learning models and produces an acceptable prediction accuracy.

Overall, our research contributed to different aspects of DH.

  • Manal Kalkatawi , King Abdulaziz University

    Manal Kalkatawi

Building 19, Hall 1 14:00 - 14:20 Details

Manal Kalkatawi , King Abdulaziz University
2:20 pm

Coffee Break

Building 19, Hall 1 14:20 - 14:35 Details

2:35 pm

The Use of Custom Embeddings Generated from Pubmed Corpora for Cancer Research

In natural language processing, one of the big questions that remain open is “what is the optimal approach to embed our natural language in a vector space?”, which essentially transforms words into series of numbers. Ideally, the numbers should represent semantic meaning. In a multidimensional space, the different dimensions should correspond to different types of meaning (e.g. size of an entity, sex of an animal) that a computer algorithm can then subsequently use to make inferences.

Big text-data endowed institutions or corporations, claim only large-sized corpora produce performant embeddings. In this presentation, we will investigate what is the minimal size of a corpus useful for extracting cancer-related statements. To this end, we developed a literature knowledge mining tool “sina” (https://github.com/dicaso/sina), that allows extracting relevant statements to specific conditions and the research question at hand, by selecting a specific corpus of documents with which to establish a custom word embedding.

Building 19, Hall 1 14:35 - 14:55 Details

Christophe Van Neste , KAUST
2:55 pm

DeepPheno: Predicting single gene knockout phenotypes

Predicting the phenotypes resulting from molecular perturbations is one of the key challenges in genetics. Both forward and reverse genetic screen are employed to
identify the molecular mechanisms underlying phenotypes and disease, and these resulted in a large number of genotype-phenotype association being available for humans and model organisms.

Combined with recent advances in machine learning, it may now be possible to predict human phenotypes resulting from particular molecular aberrations.

We developed DeepPheno, a method for predicting
the phenotypes resulting from complete loss-of-function in single genes. DeepPheno uses the functional annotations with
gene products to predict the phenotypes resulting from a
loss-of-function; additionally, we employ a two-step procedure in which we predict these functions first and then predict

This allows us to predict phenotypes associated with any known protein-coding gene. We evaluate our approach using evaluation metrics established by the CAFA challenge and compare with top performing CAFA 2 methods. Our method achieves an F_max of 0.46 which is a significant improvement over state-of-the-art F_max of 0.36.

Furthermore, we show that predictions generated by DeepPheno are applicable to predicting gene-disease associations based on comparing phenotypes, and 60% of predictions made by DeepPheno interact with a gene that is already associated with the predicted phenotype.

Building 19, Hall 1 14:55 - 15:15 Details

Maxat Kulmanov - Ph.D. student KAUST, KAUST
3:15 pm

Closing Remarks and Announcement of Poster Competition Winners

  • Prof. Takashi Gojobori, KAUST

    Prof. Takashi Gojobori

Building 19, Hall 1 15:15 - 15:45 Details

Prof. Takashi Gojobori, KAUST
7:00 pm

Dinner - Invitation Only

Dinner will be served at Pure restaurant in Island Recreation Club

Pure Restaurant 19:00 - 20:30 Details