Teaching Algorithms to Caption Ultrasound Images | Dr Mohammad Alsharid
Original Article Reference
This SciPod is a summary of the conference proceedings ‘Captioning Ultrasound Images Automatically’, from the International Conference on Medical Image Computing and Computer-Assisted Intervention. doi.org/10.1007/978-3-030-32251-9_37
About this episode
Medical professionals require years of training before they can describe ultrasound images of developing foetuses. Dr Mohammad Alsharid and colleagues from the Institute of Biomedical Engineering and Nuffield Department of Women’s and Reproductive Health at the University of Oxford suggest that this task could one day be carried out by machine learning algorithms. In their latest study, the team showed how neural networks, trained by the expert knowledge of real sonographers, could convert subtle features within the images into accurate, readable captions.
This work is licensed under a Creative Commons Attribution 4.0 International License. 
What does this mean?
Share: You can copy and redistribute the material in any medium or format
Adapt: You can change, and build upon the material for any purpose, even commercially.
Credit: You must give appropriate credit, provide a link to the license, and indicate if changes were made.
More episodes
Dr. Qiang Wang | Fishing for Findings: Uncovering the Genetics of Hearing Loss
Our hearing is amongst our most profound senses, connecting us to the surrounding world through sound. However, this connection is diminished or absent altogether in millions of people around the world because of hearing loss. Hearing loss is a common sensory disorder and is often hereditary. The condition can be caused by complex genetic factors, and so far, researchers have linked over 150 genes to hearing impairment. Now, a new collaborative study led by Dr. Qiang Wang of the South China University of Technology, Dr. Tao Cai from the National Institute of Health, and Dr. Yuan Li from the China-Japan Friendship Hospital in Beijing, has uncovered a new genetic clue, a mutation in the OXR1 gene, that could upend our understanding of hereditary hearing loss, and the eventual treatments that we develop to combat it.
Dr Ramtin Zand | A Hybrid System for Real-Time Sign Language Translation
Neuromorphic computing is a powerful tool for identifying time-varying patterns, but is often less effective than some AI-based techniques for more complex tasks. Researchers at the iCAS Lab directed by Ramtin Zand at the University of South Carolina, work on an NSF CAREER project to show how the capabilities of neuromorphic systems could be improved by blending them with specialized machine learning systems, without sacrificing their impressive energy efficiency. Using their approach, the team aims to show how the gestures of American Sign Language could be instantly translated into written and spoken language.
Dr. Sarallah Rezazadeh | Unlocking the Secrets of Aging: How Stem Cells Hold the Key to Vitality
Aging is a tale written by the cells in our bodies, although some cell types play a bigger role than others. At the crux of this story is an intriguing protagonist: the stem cell. These master builders, which can differentiate into any cell type, thereby helping to replace diseased or worn-out tissues, are essential for tissue repair and in maintaining health into old age. But as we get older, the capabilities of stem cells gradually diminish, which is known as stem cell exhaustion and is a key facet of aging itself. Stem cell exhaustion plays a role in a large number of age-related diseases, meaning that it could be a crucial research target in developing new treatments and techniques to help us age well. A Research Topic in the open-access journal Frontiers in Aging has been curated by Dr. Sarallah Rezazadeh of the Icahn School of Medicine at Mount Sinai and Professor Georgina May Ellison-Hughes of King’s College London. The Topic collects groundbreaking studies into stem cell exhaustion under one open-access roof, exploring the detailed mechanisms underlying the phenomenon and establishing the field in a wider context to identify promising therapeutic approaches for those later in life.
Dr. Mingjun Xie | A Blueprint for Life: Creating Blood Vessels in Bioprinted Tissues
A future where injured or diseased organs can be removed and replaced with new lab-printed tissues that are customized specifically for each patient is not as far away as you might think. These functional and living tissues could grow naturally within the body, and repair and sustain themselves over time. While these concepts were once in the realm of science fiction, advances in bioprinting, which is a form of 3D printing using biological “inks” (known as bioinks) loaded with living cells, are bringing them closer to reality. Among the researchers advancing this field is Dr. Mingjun Xie of Zhejiang University, China, and colleagues, who are performing work that addresses a significant challenge in bioprinting. This involves creating large portions of tissues that have a functional vasculature, thereby mimicking the complexity of native tissues and organs.
Increase the impact of your research
• Good science communication helps people make informed decisions and motivates them to take appropriate and affirmative action.
• Good science communication encourages everyday people to be scientifically literate so that they can analyse the integrity and legitimacy of information.
• Good science communication encourages people into STEM-related fields of study and employment.
• Good public science communication fosters a community around research that includes both members of the public, policymakers and scientists.
• In a recent survey, 75% of people suggested they would prefer to listen to an interesting story than read it.
Step 1 Upload your science paper
Step 2 SciPod script written
Step 3 Voice audio recorded
Step 4 SciPod published