Latest News Archive

Please select Category, Year, and then Month to display items
Previous Archive
12 May 2020 | Story Andre Damons | Photo Pexels

A data scientist and research coordinator at the University of the Free State (UFS), in collaboration with his supervisor at the University of Pretoria (UP), is at the forefront of the fight against the Covid-19 virus with accurate data and analysis.
Herkulaas Combrink of the Centre for Teaching and Learning at the UFS and PhD candidate in Computer Science at the UP, said accurate data is important to prevent widespread panic and sensationalism during a global disaster such as the current pandemic. This information helps people to make informed decisions and to reduce their exposure to the threat of the virus.

Assisting decision-makers

“I, along with colleagues from the World Health Organization, the Centers for Disease Control and Prevention in the USA, the provincial office of the Centers for Disease Control and Prevention, provincial clinicians, and the Free State Department of Health led by Dr David Motau, have been able to progress significantly in terms of evidence-based tools to assist provincial and national decision-makers during these turbulent times.”
“It does come at a cost, though, in that we have worked continuously since the lockdown, dedicating all our time and efforts to the department from all over to ensure that we are not part of some of the global statistics we have seen,” said Combrink. 

A paper written together with his supervisor, Dr Vukosi Marivate, has also been accepted by the Department of Higher Education and Training (DHET)-accredited Data Science Journal.  This paper is related to a framework for sharing public data to the public in a way that is useful, usable, and understandable. 

Ongoing projects

Combrink said it is hard to name all those who are/were involved in the great work done by the Free State Department of Health, but some of them include Dr Elizabeth Reji (Head of Department, Family Medicine), Dr Collin Noel (surgeon, senior lecturer at the UFS), Dr Sammy Mokoena (community health registrar, UFS), Dr Ming-Han Motloung (public health medicine specialist, senior lecturer, UFS), Dr Perpetual Chikobvu (Director: Information Management at the Department of Health, affiliated lecturer at the UFS), as well as Alfred Deacon (lecturer at the UFS), who have worked at some point during this short space of time on one of the many projects. 

Some of the projects include the following:

• A provincial database for screening and monitoring.
• A data pipeline and assembly of hospital information flow, liaised with the NICD, Vodacom, and the different district managers to ensure that the pipeline occurs in a timely manner.
• Digitised paper-based capturing tools for rapid data capturing and processing.
• Incorporated state-of-the-art visualisation tools to action data into useful information for decision-makers in certain areas.
• Provided both provincial and national projections, stress testing different scenarios using a variety of statistical, computational, and/or machine-learning approaches to add to the already existing projections of the Council for Scientific and Industrial Research (CSIR).
• Training healthcare professionals in the field to apply these tools within their own districts.
No easy task

“These aforementioned feats were by no means easy and are not completed yet, but we are getting there. In the foreseeable future, I will be working closely with national and international researchers to deploy a tool for hospital managers in the Free State that will assist them when we move from level 5 to any level below.”

“In addition to this, I am constantly providing support to the Free State Department of Health regarding any analysis required for decision-making purposes. The teams we work in comprise highly competent individuals with a passion for solving problems from multidisciplinary perspectives,” according to Combrink.

News Archive

Mathematical methods used to detect and classify breast cancer masses
2016-08-10

Description: Breast lesions Tags: Breast lesions

Examples of Acho’s breast mass
segmentation identification

Breast cancer is the leading cause of female mortality in developing countries. According to the World Health Organization (WHO), the low survival rates in developing countries are mainly due to the lack of early detection and adequate diagnosis programs.

Seeing the picture more clearly

Susan Acho from the University of the Free State’s Department of Medical Physics, breast cancer research focuses on using mathematical methods to delineate and classify breast masses. Advancements in medical research have led to remarkable progress in breast cancer detection, however, according to Acho, the methods of diagnosis currently available commercially, lack a detailed finesse in accurately identifying the boundaries of breast mass lesions.

Inspiration drawn from pioneer

Drawing inspiration from the Mammography Computer Aided Diagnosis Development and Implementation (CAADI) project, which was the brainchild Prof William Rae, Head of the department of Medical Physics, Acho’s MMedSc thesis titled ‘Segmentation and Quantitative Characterisation of Breast Masses Imaged using Digital Mammography’ investigates classical segmentation algorithms, texture features and classification of breast masses in mammography. It is a rare research topic in South Africa.

 Characterisation of breast masses, involves delineating and analysing the breast mass region on a mammogram in order to determine its shape, margin and texture composition. Computer-aided diagnosis (CAD) program detects the outline of the mass lesion, and uses this information together with its texture features to determine the clinical traits of the mass. CAD programs mark suspicious areas for second look or areas on a mammogram that the radiologist might have overlooked. It can act as an independent double reader of a mammogram in institutions where there is a shortage of trained mammogram readers. 

Light at the end of the tunnel

Breast cancer is one of the most common malignancies among females in South Africa. “The challenge is being able to apply these mathematical methods in the medical field to help find solutions to specific medical problems, and that’s what I hope my research will do,” she says.

By using mathematics, physics and digital imaging to understand breast masses on mammograms, her research bridges the gap between these fields to provide algorithms which are applicable in medical image interpretation.

We use cookies to make interactions with our websites and services easy and meaningful. To better understand how they are used, read more about the UFS cookie policy. By continuing to use this site you are giving us your consent to do this.

Accept