.png)
Curated by Alim Bidmus Clinical Analyst with Lagos State Primary Healthcare Board @Abdulbidmus on 26th July 2021
Every week the global OHT Twitter account is curated by a wonderful member of the OHT community. They share with us how they do what they do, what they're interested in, their top tips and general learnings. We like to turn these Tweets into blogs as there is so much goodness in them!
Go on... sign up to curate our account on a Monday soon...you know you want to!
My name is Alim Bidmus from Lagos Nigeria. I work as a Clinical Analyst with Lagos State Primary Healthcare Board and I am the founder of a HealthTech startup - MedVendorHub (A virtual platform for patients to access medical services) - medvendor.com On today's Twitter feed, I made a brief introduction about myself and my job, I discussed my recent project as a lead AI engineer for a machine learning model project developed for ophthalmic care
Good beautiful morning everyone. My name is Alim @AbdulBidmus
and I’m so thrilled to be your curator for the day. I work with @lsphcb – Lagos State Primary Healthcare Board as a Clinical Analyst and I’m the lead AI engineer of a HealthTech startup in Lagos #OHT21#TwitterTakeover
Today, i will be discussing about a recent HealthTech project with
@eyefoundationng where I deployed a Machine learning model for diagnosing ophthalmic conditions #HealthTech#MachineLearning#Ophthalmology
One of the common types of #AI is machine learning. It analyzes and discovers patterns in large data sets in order to aid decision-making. #Algorithms are the building blocks of machine learning which are a set of instructions for completing a set of tasks. #AI#ML#HealthTech
The goal of Artificial Intelligence in Healthcare isn't to replace physicians but to aid medical decisions. Physicians have a non linear method in reaching a medical diagnosis #OHT21#Physicians#Robotics#AI
One of the major challenge with the project was getting datasets of retinal fundus images of these eye conditions – diabetic retinopathy, age-related macular degeneration, glaucoma, retinopathy of prematurity, and congenital cataract Machine Learning and AI in Ophthalmology - Glaucoma Today
Because of the project's uniqueness, I couldn't rely on the vast majority of open-source datasets available online. I had to obtain data from major ophthalmology facilities in Lagos, as well as a few open sources from which I could confirm their authenticity. #AI#Datasets#OHT21
A multi-class classification method was used in grouping the data. I selected this method because we training the machine to diagnose about 5 different ophthalmic conditions #AI#MachineLearning#HealthTechnology#OHT21
I continued to get the right platform for training the model which is Tensorflow. TensorFlow is an end-to-end open source platform for machine learning #OHT21#MachineLearning#TensorFlow#HealthTech
Deploying the model was a bit tasking. Had to consider various options of storage and also cost. I went for Azure ML which is a cloud based platform used to train, deploy and automate machine learning models. Azure ML is a versatile platform as it supports Python, R, & R studios.
The ML dev cycle is in 7 stages - data sourcing, data prep, data wrangling, analyze data, train and test the model, and lastly deployment of the model #ML#Cycle#Deployment#ML#AI#OHT21
It was a great experience working as the Lead AI engineer with
@eyefoundationng in developing the ML model. In its few weeks of usage, there has been a decrease in patient's waiting time.
It's my last tweet for today and all thanks to @One_HealthTech
for letting me host #Mondaytakeover On this note I say thanks and good night