Connect with us

Technology

Faster, More Accurate Diagnoses: Healthcare Applications Of AI Research

Published

on

When Google DeepMind’s AlphaGo shockingly defeated legendary Go player Lee Sedol in 2016, the terms artificial intelligence (AI), machine learning and deep learning were propelled into the technological mainstream.

AI is generally defined as the capacity for a computer or machine to exhibit or simulate intelligent behaviour such as Tesla’s self-driving carand Apple’s digital assistant Siri. It is a thriving field and the focus of much research and investment. Machine learning is the ability of an AI system to extract information from raw data and learn to make predictions from new data.

Deep learning combines artificial intelligence with machine learning. It is concerned with algorithms inspired by the structure and function of the brain called artificial neural networks. Deep learning has received much attention lately both in the consumer world and throughout the medical community.

READ MORE | The Robots Are Coming: How To Stay Relevant And Re-Skill Yourself

Interest in deep learning surged with the success of AlexNet, a neural network designed by Alex Krizhevsky that won the 2012 ImageNet Large Scale Visual Recognition Challenge, an annual image classification competition.

Another relatively recent advancement is the use of graphical processing units (GPUs) to power deep learning algorithms. GPUs excel at computations (multiplications and additions) needed for deep learning applications, thereby lowering application processing time.

In our lab at the University of Saskatchewan we are doing interesting deep learning research related to healthcare applications — and as a professor of electrical and computer engineering, I lead the research team. When it comes to health care, using AI or machine learning to make diagnoses is new, and there has been exciting and promising progress.

Extracting blood vessels in the eye

Detecting abnormal retinal blood vessels is useful for diagnosing diabetes and heart disease. In order to provide reliable and meaningful medical interpretations, the retinal vessel must be extracted from a retinal image for reliable and meaningful interpretations.

Although manual segmentation is possible, it is a complex, time-consuming and tedious task which requires advanced professional skills.

My research team has developed a system that can segment retinal blood vessels simply by reading a raw retinal image. It is a computer-aided diagnosis system that reduces the work required by eye-care specialists and ophthalmologists, and processes images 10 times faster, while retaining high accuracy.

READ MORE | Businesses Of The Future: 20 New Wealth Creators On The African Continent

Detecting lung cancer

Computer tomography (CT) is widely used for lung cancer diagnosis. However, because visual representations of benign (non-cancerous) and malignant (cancerous) lesions in CT scans are similar, a CT scan cannot always provide a reliable diagnosis. This is true even for a thoracic radiologist with many years of experience.

The rapid growth of CT scan analysis has generated a pressing need for advanced computational tools to assist radiologists with the screening progress.

To improve radiologists’ diagnostic performance, we have proposed a deep learning solution. Based on our research findings, our solution outperforms experienced radiologists. Moreover, using a deep learning-based solution improves diagnostic performance overall and radiologists with less experience benefit from the system the most.

READ MORE | Meet The Top Women Investors Of The Midas List In 2019

Limitations and challenges

Although great promise has been shown with deep learning algorithms in a variety of tasks across radiology and medicine, these systems are far from perfect. Obtaining high-quality annotated datasets will remain a challenge for deep learning training. Most computer vision research is based on natural images, but for healthcare applications, we need large annotated medical image datasets.

Another challenge from a clinical standpoint will be the time to test how well deep learning techniques perform in contrast to human radiologists.

There needs to be more collaboration between physicians and machine learning scientists. The high degree of complexity of human physiology will also be a challenge for machine learning techniques.

Another challenge is the requirements to validate a deep learning system for clinical implementation, which would likely require multi-institutional collaboration and large datasets. Finally, an efficient hardware platform is required to ensure fast processing of deep learning systems.

In the complex world of healthcare, AI tools can support human practitioners to provide faster service and more accurate diagnoses, and analyze data to identify trends or genetic information that may predispose someone to a particular disease. When saving minutes can mean saving lives, AI and machine learning may be transformative for healthcare workers and patients.

Seokbum Ko; Professor, University of Saskatchewan

The Conversation

Continue Reading
Advertisement
Comments

Technology

Where The Medium’s The Topic And The Topic is Topical

Published

on

UJ, 4IR, and the CloudebateTM concept

UJ is the University of Johannesburg. 4IR is the Fourth Industrial Revolution. CloudebateTM? Well – it’s a place where really interesting questions are asked, such as: is the academic thesis a thing of the past? Have books outlived their physical form? Are we witnessing the demise of childhood? Will eye-tracking, sip and puff, or exoskeletons lead to true equality of opportunity? Will society change Africa? Will Africa help change society? Will education teach our children what they really need to know? And if so, how?

As 4IR sweeps the world, sending many preconceptions, predilections, and presuppositions tumbling as it goes, UJ sees the asking of questions like these as a fundamental response. And it’s responding because, since 2013, when it first embarked on its strategy of global excellence and stature, the university saw a clear need to take the lead in exploring the applications, implications and potential of 4IR. What’s more, it saw a need to do this not just as part of its positioning as a thought-leader on the continent, but as part of making a proactive and positive contribution towards African society, education and enablement.

A vision of width, a platform of depth

It’s a significant vision, and as part realising it, UJ has been investigating new and challenging ways, not just of identifying the issues at stake, but of presenting them in depth. It sought a way that would bring medium and content, idea and action, debate and initiative, together on one unique platform.

And that unique platform, one that UJ has not only created, but given a unique name to as well, is the CloudebateTM

The CloudebateTM

The CloudebateTM has essentially taken the traditional debate/panel discussion and reimagined it, placing it firmly within the realm of its own 4IR scope, and using the latest live-streaming technology. It is the place where 4IR ideas that have been identified as relevant, meaningful, challenging and thought-provoking are placed before an expert panel as well as an online audience who are invited to participate in real time, online, in a very 4IR way, in the discussion, analysis and dissection.  

There have been seven Cloudebates held so far, and their names provide an insight into their capacity to provoke thought: The Way Tomorrow Works; Digitally Equal; Is 4IR the Demise of Childhood? Questioning the Answers; Obsolete or Absolute? Should Books be Shelved? Adding Muscle to Open Doors.

When thought is action

It’s all about the kind of world we are creating for our children to inhabit. What will the elimination of jobs do to society? Are children growing directly into the immediacy of adulthood? Are academic theses outdated? Are libraries passé? Can technology enable opportunity equally for all?

The digital reach has been immense, not just in South Africa but globally, where it has found a worldwide audience. Moreover, UJ’s CloudebateTM initiative is set to continue into 2020 with further challenges to our received wisdom, our perceived way of doing things. So, if you have any stimulating 4IR topics that you would like to see discussed, send them to [email protected] – UJ would love to hear from you. And if you’d like to see the discussions that have already taken place, then just go to uj.ac.za/4IR, where you can watch, and take a view of your own.

Creating tomorrow

With its innovative CloudebateTM concept, UJ’s pursuit of global excellence has been a most rewarding journey that will continue to develop and expand along with 4IR, and along with UJ’s ongoing commitment to creating tomorrow.

Content provided by the University of Johannesburg

Continue Reading

30 under 30

Applications Open for FORBES AFRICA 30 Under 30 class of 2020

Published

on

FORBES AFRICA is on the hunt for Africans under the age of 30, who are building brands, creating jobs and transforming the continent, to join our Under 30 community for 2020.


JOHANNESBURG, 07 January 2020: Attention entrepreneurs, creatives, sport stars and technology geeks — the 2020 FORBES AFRICA Under 30 nominations are now officially open.

The FORBES AFRICA 30 Under 30 list is the most-anticipated list of game-changers on the continent and this year, we are on the hunt for 30 of Africa’s brightest achievers under the age of 30 spanning these categories: Business, Technology, Creatives and Sport.

Each year, FORBES AFRICA looks for resilient self-starters, innovators, entrepreneurs and disruptors who have the acumen to stay the course in their chosen field, come what may.

Past honorees include Sho Madjozi, Bruce Diale, Karabo Poppy, Kwesta, Nomzamo Mbatha, Burna Boy, Nthabiseng Mosia, Busi Mkhumbuzi Pooe, Henrich Akomolafe, Davido, Yemi Alade, Vere Shaba, Nasty C and WizKid.

What’s different this year is that we have whittled down the list to just 30 finalists, making the competition stiff and the vetting process even more rigorous. 

Says FORBES AFRICA’s Managing Editor, Renuka Methil: “The start of a new decade means the unraveling of fresh talent on the African continent. I can’t wait to see the potential billionaires who will land up on our desks. Our coveted sixth annual Under 30 list will herald some of the decade’s biggest names in business and life.”

If you think you have what it takes to be on this year’s list or know an entrepreneur, creative, technology entrepreneur or sports star under 30 with a proven track-record on the continent – introduce them to FORBES AFRICA by applying or submitting your nomination.

NOMINATIONS AND APPLICATIONS CRITERIA:

Business and Technology categories

  1. Must be an entrepreneur/founder aged 29 or younger on 31 March 2020
  2. Should have a legitimate REGISTERED business on the continent
  3. Business/businesses should be two years or older
  4. Nominees must have risked own money and have a social impact
  5. Must be profit generating
  6. Must employ people in Africa
  7. All applications must be in English
  8. Should be available and prepared to participate in the Under 30 Meet-Up

Sports category

  1. Must be a sports person aged 29 or younger on 31 March 2020
  2. Must be representing an African team
  3. Should have a proven track record of no less than two years
  4. Should be making significant earnings
  5. Should have some endorsement deals
  6. Entrepreneurship and social impact is a plus
  7. All applications must be in English
  8. Should be available and prepared to participate in the Under 30 Meet-Up

Creatives category

  1. Must be a creative aged 29 or younger on 31 March 2020
  2. Must be from or based in Africa
  3. Should be making significant earnings
  4. Should have a proven creative record of no less than two years
  5. Must have social influence
  6. Entrepreneurship and social impact is a plus
  7. All applications must be in English
  8. Should be available and prepared to participate in the Under 30 Meet-Up

Your entry should include:

  • Country
  • Full Names
  • Company name/Team you are applying with
  • A short motivation on why you should be on the list
  • A short profile on self and company
  • Links to published material / news clippings about nominee
  • All social media handles
  • Contact information
  • High-res images of yourself

Applications and nominations must be sent via email to FORBES AFRICA journalist and curator of the list, Karen Mwendera, on [email protected]

Nominations close on 3 February 2020.

Continue Reading

Technology

Facebook Is Still Leaking Data More Than One Year After Cambridge Analytica

Published

on

By

Facebook said late Tuesday that roughly 100 developers may have improperly accessed user data, which includes the names and profile pictures of individuals in certain Facebook Groups.

The company explained in a blog post that developers primarily of social media management and video-streaming apps retained the ability to access Facebook Group member information longer than the company intended.

The company did not detail the type of data that was improperly accessed beyond names and photos, and it did not disclose the number of users affected by the leak.

Facebook restricted its developer APIs—which provide a way for apps to interface with Facebook data—in April 2018, after the Cambridge Analytica scandal broke the month before. The goal was to reduce the way in which developers could gather large swaths of data from Facebook users.

But the company’s sweeping changes have been relatively ineffective. More than a year after the company restricted API access, the company continues to announce newly discovered data leaks.

“Although we’ve seen no evidence of abuse, we will ask them to delete any member data they may have retained and we will conduct audits to confirm that it has been deleted,” Facebook said in a statement.

The social media giant says in its announcement that it reached out to 100 developer partners who may have improperly accessed user data and says that at least 11 developer partners accessed the user data within the last 60 days.

Facebook has been reviewing the ways that companies are able to collect information and personal data about its users since the New York Times reported that political consulting firm Cambridge Analytica harvested data of millions of users. Facebook later said the firm connected to the Trump campaign may have improperly accessed data on 87 million users.

The Federal Trade Commission slapped Facebook with a $5 billion fine as a result of the breach. As part of the 20-year agreement both parties reached, Facebook now faces new guidelines for how it handles privacy leaks.

“The new framework under our agreement with the FTC means more accountability and transparency into how we build and maintain products,” Facebook’s director of platform partnerships, Konstantinos Papamiltiadis, wrote in a Facebook post.

“As we work through this process we expect to find examples like the Groups API of where we can improve; rest assured we are committed to this work and supporting the people on our platform.”

Michael Nuñez

Continue Reading

Trending