Hello, I’m a Pharmacist in Malaysia! I code to create fun things and interested in applications of data science in healthcare 🐳

Sharing my experience of building wheremymask.com during the mask crisis back in Jan-March 2020

Image for post
Image for post

Back in January 2020, COVID-19 has caused a dramatic shortage in the supply of face masks (The Mask Crisis), which is mainly explained by a surge in demand and slow down of production in China.

The demands for masks in Malaysia faced a massive spike due to the public fear of COVID-19 and it causes the price of a piece of a mask up from RM 0.20 normally to RM 2.00 (a 1000% increment in price).

Everyone is finding a piece of the mask to protect themself and their family that time, every day they go to search for masks and sanitizer in different pharmacies and find nothing. I hear complaints from my community pharmacist friends regarding the angry customers and they are helpless because they don’t have the stock or the stocks are still in shipping from the manufacturer. …


Use your python Tensorflow model in react today

Image for post
Image for post

You had trained your image classification model well in TensorFlow and you eagerly want to integrate it into production and use by your users. However, you found out the problem as below:

  • Due to privacy concern, your user doesn’t want to upload their photo into your server
  • You are poor and you don’t want to create a whole new server just to host a model
  • You need a high latency for your model

Thanks to Tensorflow.js (TF.js), the client-side solution that will solve the problems mentioned above. Now you can integrate it with your favorite javascript framework such as React.js …


Comparing NLTK and spaCy for text normalization in NLP

Image for post
Image for post

Why Text Normalization?

Most NLP tasks require us to refer to a dictionary to teach the machine the word’s context or vocabulary, it is locally to think that, the smaller the vocabulary the better the performance of our NLP task.

Every NLP pipeline needs to do text normalization. Text normalization is the process of transforming a text into a canonical (standard) form. It is one of the important steps in text preprocessing to reduce the noises generated by a single word with multiple forms. …


To explore the effect of balanced and unbalanced data in ML Model

Image for post
Image for post

Note from Towards Data Science’s editors: While we allow independent authors to publish articles in accordance with our rules and guidelines, we do not endorse each author’s contribution. You should not rely on an author’s works without seeking professional advice. See our Reader Terms for details.

Hepatitis is an inflammation of the liver. The condition can be self-limiting or can progress to fibrosis (scarring), cirrhosis, or liver cancer. Hepatitis viruses are the most common cause of hepatitis in the world but other infections, toxic substances (e.g. alcohol, certain drugs), and autoimmune diseases can also cause hepatitis. [1]

There are 5 main hepatitis viruses, referred to as types A, B, C, D, and E. These 5 types are of greatest concern because of the burden of illness and death they cause and the potential for outbreaks and epidemic spread. Types B and C lead to chronic disease in hundreds of millions of people and, together, are the most common cause of liver cirrhosis and cancer. …


Examples of how retail can incorporate the recommendation systems into their business practice.

Image for post
Image for post

Recommender system is an essential tool that can suggest personalized products or services for a given customer and directly help to increase business sales and attract more customers.

It is reported that 35% of what consumers purchase on Amazon come from product recommendation system [9].

Alibaba, China’s biggest e-commerce company revenue rose 51 percent to 93.5 billion yuan after the introduction of the new recommendation-based advertising [28].

Best Buy, reported a 23.7% increase in second quarter of 2016 after the concentration of online sales with product recommendation engine.

There have been few published studies that provide direct evidence of the value of recommender systems to e-Businesses such as, a consumer packaged goods supermarket LeShop [6] which demonstrated an increase of 0.30% of direct monthly revenue and at least by 66% and on average by 336% for indirect extra revenue. …


Image for post
Image for post

Introduction

The technology advancement in healthcare industry have helped us generate more and more data, it was estimate that global healthcare data generated in 2020 was 2,314 exabytes which increased from 153 exabytes in 153 exabytes [1]. These data were generated from multiple sources such as EHR (Electronic Health Records), clinical research, patient’s internet of thing (IoT), outpatient’s medical records, patient’s telehealth records and chat history, a large portion of these data takes the form of unstructured narrative text which describe the medical procedures such as clinical records, discharge summaries, clinical monitoring sheets or radiological reports [2].

These unstructured narrative texts are written by the doctors, nurses, pharmacist and staff providing care to a patient, and offer increased detail beyond the traditional discharge summary, these notes are generated during the course of care, and possesses detailed information such as the progress of a patient’s condition, the plan of care, medical and family history, as well as a number of other clinical attributes. …


Image for post
Image for post

In March 2018, Google announced Tensorflow.js version 1.0, which enable developer to use machine learning (ML) models directly in browser or in Node.js via JavaScript.

Since then I always want to use those fancy pre-trained model published by google in my project such as object detection, pose estimation and etc. It is however, those example and demo are shown in pure JavaScript and there are a less guide on how to use it in a react application.

Inspired by Benson Rua code in Tensorflow.js Community “Show & Tell”. I had decided to write something on using create-react-app + tensorflow.js …


Image for post
Image for post

In every investment decision on stock, as a rational investor, the attention is always aimed at the return of the investment. Investors will choose investments in stocks that give the highest expected return (return). However, in real life investments, each selection of stock will contain an element of uncertainty, so investors should consider the risk factor [1]. This leads to the combination of multiple stocks and the weight of holding it will result in different expected returns and different risk associated with the investors.


Image for post
Image for post

It’s my honor to be invited as a guest speaker in Malaysian Innovative Healthcare Symposium- MIHS 3.0 on 6th of October. My topic was “Coding in the time of Healthcare Revolution” and i’m sharing on how learn to code will actually benefit Healthcare Professionals in their practice. I had shown the audience how to code a Basic Medical Educational VR Application in part of my talk.

As an introduction, Virtual Reality (VR) is an emerging trend in medical world, it offers immersive experience to the users, which mean it can distract the user from outside world, and dive them into the virtual world. …

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store