基本信息
浏览量:2
职业迁徙
个人简介
Research interests
While the human brain remains unparalleled in its ability to perform highly sophisticated information processing on extremely limited energy resources, Dr Omid Kavehei's research in the field of nanoelectronics aims to lead to the development of a new breed of tiny intelligent devices to collectively emulate this capability in hardware more closely than ever. Such devices are expected to be able to solve a wide range of issues involving sensory perception.
"A healthy human brain delivers an impressive information processing capability on very limited energy resources. Unlike modern artificial intelligence (AI) models, the brain does not require excessively high amounts of data to achieve an optimum learning point. It can learn 'on the fly' from experience, memory and multisensory data streams, using input from our auditory, visual, olfactory, gustatory and tactile senses.
"Despite massive improvement in today's AI models and their successful practicality, they are still far from being truly intelligent or closely brain-inspired. They require excessively large training datasets and a form of supervised approach, and they consume a lot of energy.
"I believe low-power nanoelectronic devices that mimic operation of biological neurons and synapses could lead to truly intelligent brain-inspired systems that could learn from experience.
"Tiny intelligent devices that can run on batteries could revolutionise several industries, from medical devices to the internet of things. For instance, one potential application is epileptic seizures forecasting. Most people who are diagnosed with refractory epilepsy are not responsive to medication. Other than surgery, a possible step forward to improve these people's lives could be the development of an intelligent system that forecasts seizures using brain signals. Such a system must be ultra-low-power, run on batteries and learn from patients' neural activity over time to be truly effective. The system could either warn the patient about the possibility of an upcoming seizure or, as part of a responsive implant, activate neural stimulation to avoid seizures.
"I'm passionate about this research and its ever-growing impact on our quality of life. It's an exciting era of increasing demand for autonomous decision making of any kind. This is also a very exciting area of research as it offers unrivalled solutions to problems that are fundamentally bound by strict energy limitations. I was fortunate to be part of impactful national flagship projects like the Australian Bionic Eye. It's fulfilling to be able to give back to the community and improve people's lives.
"I've been researching in the field of nano- and microelectronics since 2003, and in this specific field since 2009. I joined the University of Sydney in 2017, where I hope to address many fascinating challenges with the support of the University's outstanding cross-disciplinary research environment and resources."
While the human brain remains unparalleled in its ability to perform highly sophisticated information processing on extremely limited energy resources, Dr Omid Kavehei's research in the field of nanoelectronics aims to lead to the development of a new breed of tiny intelligent devices to collectively emulate this capability in hardware more closely than ever. Such devices are expected to be able to solve a wide range of issues involving sensory perception.
"A healthy human brain delivers an impressive information processing capability on very limited energy resources. Unlike modern artificial intelligence (AI) models, the brain does not require excessively high amounts of data to achieve an optimum learning point. It can learn 'on the fly' from experience, memory and multisensory data streams, using input from our auditory, visual, olfactory, gustatory and tactile senses.
"Despite massive improvement in today's AI models and their successful practicality, they are still far from being truly intelligent or closely brain-inspired. They require excessively large training datasets and a form of supervised approach, and they consume a lot of energy.
"I believe low-power nanoelectronic devices that mimic operation of biological neurons and synapses could lead to truly intelligent brain-inspired systems that could learn from experience.
"Tiny intelligent devices that can run on batteries could revolutionise several industries, from medical devices to the internet of things. For instance, one potential application is epileptic seizures forecasting. Most people who are diagnosed with refractory epilepsy are not responsive to medication. Other than surgery, a possible step forward to improve these people's lives could be the development of an intelligent system that forecasts seizures using brain signals. Such a system must be ultra-low-power, run on batteries and learn from patients' neural activity over time to be truly effective. The system could either warn the patient about the possibility of an upcoming seizure or, as part of a responsive implant, activate neural stimulation to avoid seizures.
"I'm passionate about this research and its ever-growing impact on our quality of life. It's an exciting era of increasing demand for autonomous decision making of any kind. This is also a very exciting area of research as it offers unrivalled solutions to problems that are fundamentally bound by strict energy limitations. I was fortunate to be part of impactful national flagship projects like the Australian Bionic Eye. It's fulfilling to be able to give back to the community and improve people's lives.
"I've been researching in the field of nano- and microelectronics since 2003, and in this specific field since 2009. I joined the University of Sydney in 2017, where I hope to address many fascinating challenges with the support of the University's outstanding cross-disciplinary research environment and resources."
研究兴趣
论文共 29 篇作者统计合作学者相似作者
按年份排序按引用量排序主题筛选期刊级别筛选合作者筛选合作机构筛选
时间
引用量
主题
期刊级别
合作者
合作机构
medrxiv(2023)
2022 8th International Conference on Event-Based Control, Communication, and Signal Processing (EBCCSP)pp.1-5, (2022)
FRONTIERS IN NEUROLOGY (2022)
bioRxiv (Cold Spring Harbor Laboratory) (2021)
加载更多
作者统计
合作学者
合作机构
D-Core
- 合作者
- 学生
- 导师
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn