Aung Thura Htoo

[Aung Thura Htoo]

I think for me, by combining following variables would be able to recognize me.
Sex: male
Appearance: long hair
Study Major: Biomedical and Health Informatics
Study place: Faculty of Tropical Medicine, Mahidol University
Place of residence: Myanmar (Burma)

[Aung Thura Htoo]

To understand why they are not using bed nets in more detail, I will do a follow-up qualitative study on ‘barriers against bed nets’. Qualitative method is a good fit for this type of study, that tries to uncover on the social, behavioral, and physical factors influencing on the barriers against using bed nets. In the study, key informant interviews (KII) approach will be used as the main methodology. Key informants such as village heads, community health workers, and community-based organizations will be interviewed to explore the common barriers (such as cost, usability, and so on). The results from the KII will be triangulated by using focus-group discussion. By combining the two approaches, the findings from this study can provide more in-depth understanding of the reasons or the factors hindering the use of bed nets.

[Aung Thura Htoo]

External variables that might influence an individuals’ perceived ease of use or perceived usefulness of a new technology are:

1. Digital Literacy – (people with higher digital literacy might be more confident in perceived ease of use)

2. Peer Use or Pressure (Co-worker effect – social norm) – (people who see their co-workers using the system and performing well might think of perceive ease of use or perceived usefulness)

3. Adequate Support (Training and Manuals) — (people who have adequate training and support might feel more perceived ease of use)

4. Output Quality – (people who see and understand the quality output of a new technology can influence its perceived usefulness)

5. Prior Experience or Use – (people who has used similar technology before might find it easier to use – perceived ease of use).

[Aung Thura Htoo]

Hello Cing, thank you for your presentation and the demo of what blockchain is.

Q1: As discussed in the seminar, I found it hard to comply with the right to be forgotten when using blockchain technology. However, I believe that being able to strictly control whom to give access to would be a great step towards the regulation.

Q2: I believe that the patient should control their own digital health identity instead of the hospitals. Even if the hospitals control the data, the patient should have access to the exact copy of their health identity and be able to control the access.

[Aung Thura Htoo]

Efficacy, effectiveness, and efficiency are important outcomes in m experimental studies.

Efficacy: It answers how much that intervention produces an effect under controlled environments and circumstances. E.g. Testing a vaccine in a laboratory conditions (90% efficacy)

Effectiveness: It answers how much that intervention produces an effect in the real-world settings. E.g. Studying the effectiveness of the vaccine in a randomized controlled trial study (60-80 effective)

Efficiency: It measures the relation between the output of the intervention and the resources that are necessary to provide the intervention. E.g. How many cases of vaccine preventable diseases averted per cost

[Aung Thura Htoo]

Hello, I think ‘digital literacy’ could be a potential confounder in this case. It satisfies all three criteria of a confounder. Firstly, it is associated with the exposure (age): younger adults are more likely to have higher digital literacy compared to older adults. Secondly, it is associated with the outcome (active contact pattern): people who have higher digital literacy are more likely to use the mobile applications and record their contacts more accurately, resulting in more observed contact patterns. Thirdly, it is not an intermediate variable in the causal pathway. So, for the above reasons, I believe that ‘digital literacy’ could be a potential valid confounder in this case.

[Aung Thura Htoo]

1. Case-Fatality Rate (CFR)

Definition: The case-fatality rate measures the proportion of individuals diagnosed with a particular disease who died from that disease within a specific time frame.
Formula: CFR = (number of deaths from a specific disease/total number of diagnosed cases) x 100
Usefulness: It can used to assess the severity or lethality of a disease among those who are affected (diagnosed). For example, a disease with a high CFR indicates a more lethal or severe one.

2. Maternal Mortality Rate (MMR)

Definition: The maternal mortality rate is a measure of the annual number of maternal deaths per 100,000 live births. A maternal death is defined as the death of a woman while pregnant or within 42 days of termination of pregnancy, from any cause related or aggravated by the pregnancy or its management (but not from accidents).
Formula: MMR = (number of maternal deaths in a given period/total number of live births in the same period) x100,000
Usefulness: It can be used to indicate the quality of health system especially reproductive health services.

3. Age-Specific Mortality Rate (ASMR)

Definition: The age-specific mortality rate (ASMR) is a measure of number of deaths among a specific age group within a specified period per 1,000 people in that same age group during a given period.
Formula: (Number of deaths in a specific age group during a year/Mid-year population of the same age group) x1,000
Usefulness: It can be used to identify high-risk age group to provide targeted interventions and allocate resources accordingly.

[Aung Thura Htoo]

Hello Ajarn, I am sorry that I could not make the meeting on Thursday morning. My question is: when the dataset is small, first use cross validation (CV), then train the model on split train data from CV, and test the model on split test data from CV?

Then when the model is large, split the data into train and test data, and train and test accordingly.

I am a little bit confused about the role of cross validation in both training and testing phase.

[Aung Thura Htoo]

Since my model is the population growth model of Anopheles, my consideration for interventions would be ‘Indoor Residual Spraying’ (IRS). It involves spraying of mosquito resting places such as doors and ceilings with long-lasting (residual) insecticides. It aims to kill adult mosquito. So, in my compartmental model, it will target on adult mosquito compartment as extra effects of mortality on adults due to IRS. However, since my model has coupled from one stage to another. It will take effect on other compartments as well. Biologically, it makes sense. Since the fewer the numbers of adult female mosquitoes, the fewer the number of eggs and so on.
There are things to consider in my intervention formula such as coverage and efficacy. According to US CDC (2024), ideally, the coverage should be 80% to be effective but, in my model, it would be only 60% effective (since 80% is an ideal number). Additionally, the efficacy at the start is high according to Dengela et al., 2018. So, in my model, I will use 50% maximum efficacy at the start. And the residual effect lasts 4 to 6 months according to Dengela et al., 2008. So, in my model, I will use residual half-life as 2 months (60 days). By using these three parameters (coverage:60%, maximum efficacy at the start:50%, and residual half-life: 60 days), I will model the added mortality for the adult female mosquitoes in my model by starting the IRS campaign at the day 30.

References
Centers for Disease Control and Prevention. (2024, April 1). Indoor Residual Spraying Prevention Strategies. National Center for Emerging and Zoonotic Infectious Diseases. https://www.cdc.gov/malaria/php/public-health-strategy/irs-strategies.html
Dengela, D., Seyoum, A., Lucas, B., Johns, B., George, K., Belemvire, A., Caranci, A., Norris, L. C., & Fornadel, C. M. (2018). Multi-country assessment of residual bio-efficacy of insecticides used for indoor residual spraying in malaria control on different surface types: results from program monitoring in 17 PMI/USAID-supported IRS countries. Parasites & vectors, 11(1), 71. https://doi.org/10.1186/s13071-017-2608-4

[Aung Thura Htoo]

The model that I would like to explore is the temperature dependent ‘Population Growth Model’ and I have started adjusting my R codes for population growth using ‘desolve’ package. The overview of my model comes from Gizaw et al., 2025. The model has 7 compartments: Egg stage, four larval stage, pupal stage, and adult (female) state.
The key characteristics that I need to consider are the ‘mortality’ function (mu) and ‘transition’ function (beta) from one phase to another. Additionally, temperature-dependent the oviposition rate as well as biting rate are also included in this model.
As described in the picture from Gi Zaw et al., 2025, the arrow that contains (t) are considered as temperature-dependent term. So, respective temperature dependent functions will be obtained from the literature.

Source of Parameters:
1. Temperature-dependent mortality and transition functions will be used from Gizaw et al., 2025.
2. Temperature-dependent oviposition and biting rate functions will be used from Traoré et al., 2021.
Limitations: The carrying capacity for each stage cannot find, so I am considering exponential growth instead of logistic. And I will stimulate the temperature that are similar to early rainy season: providing enough resources to grow.

References:
Traoré, B., Barro, M., Sangaré, B. & Traoré, S. (2021). A temperature-dependent mathematical model of malaria transmission with stage-structured mosquito population dynamics. Nonautonomous Dynamical Systems, 8(1), 267-296. https://doi.org/10.1515/msds-2020-0138
Gizaw, A. K., Erena, T., Simma, E. A., Menbiko, D. K., Etefa, D. T., Yewhalaw, D., & Deressa, C. T. (2025). Modeling the variability of temperature on the population dynamics of Anopheles arabiensis. BMC Research Notes, 18, Article 132. https://doi.org/10.1186/s13104-025-07153-y

[Aung Thura Htoo]

I am interested in researching the growth of Anopheles mosquitoes at a village level. Anopheles mosquitoes are responsible for transmission of malaria by carrying malaria parasites such as Plasmodium falciparum and vivax.
I am trying to understand the impact of local temperature on the growth of Anopheles vector (egg, larva, pupae, and adult). Temperature is one of the parameters that influences the growth of mosquito in their life stages. Research question would be: ‘What is the impact of local temperature on the growth of Anopheles population?’
Understanding the impact of local temperature on the growth can be useful in predicting the high-risk and low-risk areas, which in turn can be used by policymakers in tailored intervention and control strategies. This is currently my research topic for my independent study (more focus on applying the model rather than formulating the whole model from scratch).
I understand that the modeling the temperature dependent population growth could be complex. So, please feel free to share your comments and suggestions.

[Aung Thura Htoo]

I have faced similar issues when running the code. So, the issue mostly is some of the library used in the codes are archived. So, I need to install those packages from elsewhere. I think there are two or three packages that cannot be installed easily. Mostly I solved it by searching those packages in CRAN repositories and tried to install them. It worked for me.

[Aung Thura Htoo]

1. The author stated that there has been an increase in the number of suicides in Thailand, higher than 6 suicides per 100,000 inhabitants. Additionally, Thailand is different from other countries in terms of economic and social factors. Besides, there is a lack of study using macro-level data on this topic.

2. One potential risk factor is the prevalence of drinking. In the study, the rate of suicide increases with the increase in the prevalence of drinking. The author stated that alcohol reduces self-control and encourages people with severe mental health problems like depression to commit suicide. I agree with his discussion. Alcohol reduces one’s ability to think clearly and, most of the time, even motivates one to perform harmful actions.

3. Statistical modeling quantifies the relationship between social, economic, and other relevant factors and the rate of suicide using provincial data. Without the use of statistical method, one would assume that the economic hardship and lower income would lead to higher suicide rate. However, the result of the study using statistical model shows that it is not true by providing the estimate of coefficients and its direction.
Additionally, using regression, the author clearly demonstrated the spatial aspects of Thai suicide problem. For example, provinces with higher rate of divorce can tend to increase the suicide rate if other variables are held constant. It can assist in allocating necessary resources and policies according to the predicted rate of each province.

[Aung Thura Htoo]

1. There are many established public health studies on the comparison between time and person. However, the comparison between small-scale location study have not been incorporated as much as other components. The reasons are due to the availability and quality of data, lack of advanced software to visualize and analyze spatial data, the complexities of analytical framework, and the privacy and confidentiality issues.
1. Spatial epidemiology can be considered as an interdisciplinary science because it incorporates principles and concepts from other fields to have a better understanding of the distribution and patterns of public health concern across different locations. For example, it incorporates with the fields of statistics, environmental science, social science.
2. The place where we live and work can be regarded as a potential disease determinant because we are exposed to the things that are attached to those places for a significant portion of our life. For example, the places we work might be an area where the pollution is significantly higher than other places, resulting in higher morbidity rate. Additionally, the places we live might be near to vegetation and grasses, leading to higher rate of vector-borne diseases.

[Aung Thura Htoo]

1. What are the user ratings for different vaccination apps?

According to the study, most apps included in this systematic review has a user rating of 4-5. It means that users are mostly satisfied with such vaccination related apps.

3. What factors influence user satisfaction with vaccination apps?

Content, functionality, experience, privacy, and service attitude are the five main areas that are related to the user satisfaction with vaccination apps. Among them, users were dissatisfied with the privacy issues. Additionally, users experience on independence apps were better due to the nature of strict review of such apps and the information being updated regularly.

4. What role does privacy play in user evaluations?

Privacy plays an important role in user evaluations because in health applications, users are usually required to put some personal health information such as their vaccination records, or vital signs or so on. So, it is paramount important that their health data are handled in accordance with confidentiality and privacy. In the study, privacy is the most dissatisfied area in user evaluation. It highlights the need to improvement in privacy aspects of vaccination apps.

[Aung Thura Htoo]

1. Intercountry disease transmission and inclusion of broader population characteristics in tracking mobility should be potential follow-up studies according to the article.

2. Government restrictions during COVID-19 in Thailand have shown to be effective in restricting the mobility of people in the villages near the border. Before any restriction, people in those villages make more long trips to the neighboring country, compared to short trips. After restriction, people make more short trips than long trips. Additionally, curfews are also proved to be effective in restricting the mobility. RoG value decreased significantly after imposing the nighttime curfew.

3. After restrictions eased, the mobility pattern reaches almost back to normal, highlighting the need of caution in relaxing the restriction so as to deter second wave or another potential episode.

[Aung Thura Htoo]

Hello Ajarn, thank you for clarifying the role of cross validation.

[Aung Thura Htoo]

Hello Aye, thank you for sharing your model and explanation. I believe adding vaccination as a compartment into your existing model of SIR, can provide more comprehensive views on how they interplay. How and where do you plan to find the parameters necessary for your vaccination compartment?

[Aung Thura Htoo]

Hello Tanaphum, thank you for sharing your intervention of choice for your model. Having a vaccination class can add more understanding on how the population in your SIRS model.

[Aung Thura Htoo]

Dear Cing, thank you for your great question. In my model, I used residual half-life instead of overall duration of residual effect so that I can model the continuous change by considering the decaying rate as residual half-life.

[Aung Thura Htoo]

Hello Wannisa, thank you for sharing your intervention of choice. I believe both interventions are relevant to your model of interest. Additionally, not only the efficacy of the treatment, but also the coverage and how well the patients adhere to standard treatments can also affect your model.

[Aung Thura Htoo]

Hello Wannisa, yes, it could be better if seasonal patterns can be incorporated into the model. However, my model only accounts for temperature, so it can only show seasonal patterns of mosquito population dynamics affected by temperature.

[Aung Thura Htoo]

Hello Cing, thank you for sharing your interesting model. You have used incubation period in your model, most of the time, latent period is used when model the change from the exposed to infectious stage. I believe latent period can be longer or shorter than incubation period depending on the nature of the infection. A person can be asymptomatic as well as infectious. Any thoughts on using incubation period in your model?

[Aung Thura Htoo]

Hello Alex, thank you for sharing your model of interest. Malaria has been one of the persistent infectious disease in the rural and hilly areas. It would be great to see how the model plays out. My question is since malaria is vector-borne disease, will you model the transmission link between human and vector?

[Aung Thura Htoo]

Hello Wannisa, thank you for sharing your modeling topic. I find it quite interesting. By any chance, do you have any flow diagram of your transmission model? Would it be coupled between compartments as you consider co-infections with TB?

[Aung Thura Htoo]

Hello Wannisa, thank you for your questions. The biting rate is also temperature dependent in the model. The function will be extracted from one of the references (Traoré et al., 2021). I will only need the value of temperature at that period to calculate the biting rate.

[Aung Thura Htoo]

Hello Wannisa, thank you for sharing your research. I think it will be an interesting to model the growth of the patients with particular diseases. However, I believe that challenge would be estimating the parameters influencing the growth or decline of the patients. I hope you could find useful parameters for your model in the existing literature.

[Aung Thura Htoo]

Hello Tanaphum, thank you for sharing your project. It sounds like an interesting study. And I am also interested in how computational methods can be used in estimating parameters like transmission and recovery rate. The challenging issue would be estimating from different diseases as they have different rates.

[Aung Thura Htoo]

Dear Ajan Pan, thank you for your comments. Yes, it is challenging to find the parameters needed for the model but I found some interesting articles that might provide most of the parameters. However, I still might need to make some assumptions for a few missing parameters.

[Aung Thura Htoo]

Dear Than Soe Oo, thank you for sharing your discussion. As you have mentioned, the limited number of study in the Thailand context is one of the main reason why this research was conducted. Additionally, I agree with you that spatial analysis can assist policy makers in identifying high risk areas and implementing targeted interventions as well as allocation of necessary resources in much needed provinces.

[Aung Thura Htoo]

Dear Wannisa, thank you for sharing your discussion. I agree with you that statistical analysis such as regression models are powerful in their ability to identify the direction of the relationship between predictors and outcome variables. This in turn can be beneficial in deciding intervention measures.

[Aung Thura Htoo]

Hello Thinzar, these are the packages that I needed to install on top of the provided codes.
install.packages(“maptools”, repos = “https://packagemanager.posit.co/cran/2023-10-13&#8221😉 (please verify the url again whether it is still there)
library(maptools)
install.packages(“fmesher”)
install.packages(“lattice”)
library(lattice)
After installing these packages, it worked for me. (I am not sure that it would work for you)
I loaded tidyverse in my workspace before, so I needed to clear my workspace again as tidyverse has some conflicts with lattice.
So, I think updating your program to latest version as suggested by Wannisa could be a great idea. Then, clear your workspace if you have any, and install and load the needed packages.

[Aung Thura Htoo]

Hello Wannisa, thank you for sharing your insights on this week’s discussion. Yes, I agree with you that it requires both skills and knowledge in order to comprehend and analyze spatial data in relation to health and disorders. Additionally, like you mentioned, the availability of database and accessible software poses as significant barriers in the past in the field of spatial epidemiology.

[Aung Thura Htoo]

Hello, thank you for sharing your discussion. I believe that the availability of quality public health data that can be linked to specific locations are still limited in many countries around the world. Even though some software and advanced statistical tests have been developed and accessible, the availability will still be one of the most challenging aspect in spatial analysis.

[Aung Thura Htoo]

May I post a follow-up question? So, the direction would be when someone has dengue, the body will response with fever but the arrow is from fever (number of days having fever) to dengue in the network. In that sense, can I assume that fever-day is the indicator of dengue outcome?

[Aung Thura Htoo]

Hello Chanapong, I agree with you that altering medical records (including vaccination) without any notification to the users can lead to health related consequences. Additionally, most apps are rated very low in the privacy aspect that highlights the need to ensure users that those apps follow their privacy policy or agreements. Thank you for sharing your discussion.

[Aung Thura Htoo]

Hello Sirithep, thank you for sharing your answers. I agree with you that developers should find ways to improve engagement with the users in a privacy respected ways, for example, making easier to report issues in the play store or app store, and taking heed of those comments and suggestions.

[Aung Thura Htoo]

Hello Ajarn, thank you for your suggestion. I think the folder structure has been the same in my file, so I will try using ‘relative’ path option.

[Aung Thura Htoo]

Hello Cing, thank you for sharing your insights. I agree with you that receiving glitchy SMS notifications can sometimes be acceptable if the frequency is not more than one. Repeatedly receiving glitchy SMS can lead to user’s dissatisfaction, like in this study. For the third issue, alteration of designated vaccination locations can be acceptable if the change is notified in advance (reasonable interval between notification and next dose date). If not, it would be hard to trust the provider who changes things without the user’s knowledge.

[Aung Thura Htoo]

Hello Ajarn Patiwat, thank you for clarifying the issue and valuable suggestion. I will try saving .qgs file before importing it into macOS.

[Aung Thura Htoo]

Hello Sirithep, thank you for sharing your insights. Yes, it is interesting how the movement pattern rebounds after relaxing the restriction of mobility. It makes me think that easing the restriction step by step would have been a better strategy so that we could still manage to contain the outbreaks if the second or another wave hits.

[Aung Thura Htoo]

Hello Phyo, thank you for sharing your answers. I agree with you that follow-up studies on the records of hospitalization or visit to a clinic would further provide additional insights regarding to disease transmission in relation with the patterns of mobility. I think the researcher could pin point the location of known clinics and hospitals in those areas and study how frequently the study participants visited those areas.

[Aung Thura Htoo]

Hello Admin, thanks a lot for your prompt response and valuable suggestion. I uploaded the content to the google drive as per your suggestion. Much appreciated.

[Aung Thura Htoo]

Hello Cing, yes, in our country, the nighttime curfews could lead to ‘near-zero’ level of mobility as most would not dare to go out even if there is a health emergency due to the fear of enforced disappearance. I totally understand your statement, we are traumatized by the word ‘curfew’.

[Aung Thura Htoo]

Hello Cing, I agree with you that restriction from government reduces the mobility across the border as well as near the border villages. Additionally, nighttime curfews are effective as well. Since the study population included both host and neighboring countries. These results are significant in studying mobility across the border in my opinion.

[Aung Thura Htoo]

Hello Chanapong, yes, I agree with you that usability is the major impact factor in terms of user engagement. Web-based platforms are more user-friendly than mobile-based counterparts, which makes web-based platforms receive more feedbacks and submission. Additionally, factors like the need to install and register are negative in user engagement. Thank you for sharing your insights.

[Aung Thura Htoo]

Hello Sirithep, thank you for sharing your answers with us. I agree with you that complexity and failure to tailor to the needs of the target audience are the negative factors in completion. Additionally, I believe the need to install an application makes it more harder to complete it as some might find it hard to install an app on their mobile for a survey.

[Aung Thura Htoo]

Yes, installing an app just to complete a survey is still a barrier for me as well. However, if it links with my other medical records, it could be worthwhile.

[Aung Thura Htoo]

Hello Cing, yes, I agree with you on the facts that the complexity in the design and the involvement of multiple steps are the major hinderance influencing on the design of mobile-based applications. Additionally, like you mentioned, digital literacy also plays an important role. As mobile apps tend to capture more detailed information, people with poor digital literacy felt discomfort using it.

[Aung Thura Htoo]

Hello Cing, yes, it is surprising to me to see that those correlation are not statistically significant. Like in the discussion, we need to determine what factors are causing it and try to fix it so that we can benefit from those technology and innovations.

[Author]

Posts