There are at least a dozen smartphone apps that can track your diet, help measure your blood sugar levels, tell you when to take your medication – there are even apps that can estimate a clinical diagnosis for your cold-like symptoms.
But what if your phone could help make life-saving discoveries while it was charging at night?
Researchers at Imperial College London used the computing power of over 83,000 mobile devices to help run anti-cancer research while they were charging. This novel approach to ‘citizen science’ helped researchers from DreamLab make two important discoveries in the space of just a few months, cutting down the time it would have taken traditional computers by decades.
The inefficient use of medical data costs billions of dollars a year in inefficient treatments or misdiagnosis. In the USA, for example, 80% of healthcare data that could be used to guide clinical treatments remains untapped meaning there are many missed opportunities to improve diagnosis and disease management. The UK’s National Health Service, for example, does not have enough skilled analysts to make sense of the information that is being collected.
“The long-term hope is to give patients a whole diet and drug arsenal in their fight against cancer and extend the project to combat other life-threatening diseases facing humanity such as cardiovascular, metabolic and neuro-degenerative disorders,” Dr Kirill Veselkov, Assistant Professor in Computational Medicine at Imperial College, told ITU.
“We are currently generating huge volumes of health data around the world every day, but just a fraction of this is being put to use. By harnessing the processing power of thousands of smartphones and AI technologies, we can tap into this invaluable resource and look for clues in the datasets,” Dr Kirill Veselkov, Assistant Professor in Computational Medicine at Imperial College, told ITU.
DreamLab, a free app from the Vodafone Foundation, uses the untapped processing power of idle smartphones – i.e. when the user is asleep at night – to fast track cancer research using a built-for-purpose platform.
“To use the DreamLab platform we needed to develop bespoke machine learning/AI code, data processing, as well as input and output data formats tuned to the specifications of the platform from scratch and without the use of third-party libraries,” Dr Veselkov explained.
Large molecular datasets were split and encoded into small data chunks – up to 2Mb in size – across millions of data files which could be distributed and processed by the DreamLab app on thousands of smartphones at the same time.
“We had to design it so that no more than 5 Mb of data is exchanged per job, memory usage is kept at ~200 Mb and a single job would run for no longer than 1-2 hours on an average smartphone,” said Dr Veselkov.
The app downloads pre-determined research problems onto user’s phones. The phones then calculate these problems using the bespoke DreamLab algorithm, and then send the result back to the research team at Imperial College.
With this research, the team was able to make two important discoveries: first, it helped to identify anti-cancer molecules in everyday foods including oranges, cabbages and grapes; secondly, that existing anti-diabetic and anti-microbial drugs could potentially be re-purposed to fight cancer.
These discoveries were made in just a few months; 100,000 smartphones running six hours a night could process the required data in three months – whereas a personal computer running 24-hours a day would take 300 years.
Speeding up the research process by such a wide margin has obvious clinical benefits for future and ongoing projects across the medical spectrum.
“The future applications of the technology within the project are to narrow down the specific food and drug combinations that can target specific cancers and their genetic networks,” said Dr Veselkov.
“The long-term hope is to give patients a whole diet and drug arsenal in their fight against cancer and extend the project to combat other life-threatening diseases facing humanity such as cardiovascular, metabolic and neuro-degenerative disorders.”
The deployment of mHealth applications and technologies has increased rapidly in recent years, with the global mHealth app market alone estimated to be worth USD 111.1 billion by 2025.
mHealth and telemedicine solutions are also expanding the availability of medical care in developing countries, with innovative solutions in care of premature babies, diagnosis of visual impairment and vaccination support already helping to save lives.
“As technology becomes more ubiquitous worldwide, we will continue to see innovative use cases in the medical field. The rapid and widespread use of technology will soon mean that everyone will have access to a doctor,” said Hani Eskandar, ICT Applications coordinator at ITU.
“The future of health is in your pocket (or on your wrist) – it is continuous and cloud connected,” Alyssa Jones, global manager of Vodafone Foundation’s DreamLab project.
DreamLab is now in four markets – Australia, New Zealand, UK and Italy – with more markets launching from August 2019 onwards.