AI in the Healthcare World
Written by Deepa Naik
Amidst the great debate of the benefits vs. harm brought in by AI related technologies (including the bone of contention between the stalwarts Zuckerberg and Musk), one industry that holds the promise of huge benefits to mankind is the Healthcare Sector.
The benefits that AI brings to the table in this sector is just too huge and very promising — be it AI virtual assistants for physicians, robotic nurses, predictive diagnostic, robotic medical imaging and data analysis or AI based research in oncology and cardiology. No wonder currently, Healthcare is one of the largest sectors to receive venture funding for AI related start-ups. Here is an interesting list which drives home this point Top 106 AI healthcare start-ups ( source: cbinsights).
The writing’s on the wall — AI is at the threshold of disrupting the healthcare world.
The fusion of technologies blurring the lines between the physical, digital, and biological spheres in what is deemed to be the start of the fourth industrial revolution is more visible in HealthCare than anywhere else. For example, here are three rather interesting stories — in the soft miniaturized robots space — a softbot with a caterpillar movement to provide medication to internal organs, a soft robotic arm inspired by octopus movement to help in surgery, a soft bot made of hydrogel to deliver drugs under your skin. This work on soft miniaturized robots is still in R&D and will be in actual use maybe a few years down the line.
However, here are some areas which are ripe and ready to reap benefits of AI.
Areas of healthcare which are seeing a lot of traction:
Patient Health Record Management
We are in the age of data explosion with a lot of information at our fingertips — including a treasure of information on patient health. Companies are using algorithms in order to analyse this data, address health challenges, and make more informed-accurate-quicker decisions. According to Technologist and investor Vinod Khosla, AI tools / Virtual Assistants will save up to 80 percent of what human physicians currently do, providing on-going support and recommendations for diagnosis and treatment, as well as administrative support, allowing physicians to focus their time on the really important elements of patient physician interaction. Apart from this, there is a lot on insights into preventive health, what makes people healthy, experimenting with disease monitoring etc.
Medical Imaging and Processing
Artificial Intelligence is changing Medical Imaging. With the ever increasing power of deep learning technologies, specifically, its power at image recognition to analyse the data collected from radiology images and applying it in unique medical cases is proving to be very beneficial for the medical imaging industry. AI algorithms read medical images similar to radiologists, by identifying patterns. AI systems are trained using vast numbers of exams to determine what normal anatomy looks like on scans from CT, magnetic resonance imaging (MRI), Ultrasound or Nuclear Imaging. Then abnormal cases are used to train the eye of the AI system to identify anomalies.
For oncology, where time plays a critical factor and early detection can be a life saver, AI based tumour imaging analysis can significantly reduce the time involved with tumour tracking assessment and speed up workflow.
Arterys, a medical imaging cloud platform is designed to acquire seven dimensions of cardiac related data including 3D heart anatomy, blood flow rate, and blood flow direction to reform cardiac MRI.
Accelerated Drug Development and Clinical Trials
Developing pharmaceuticals through clinical trials is a long process sometimes taking more than a decade and costs billions of dollars. Drug discovery using artificial intelligence and machine learning is helping cut down on the time and costs and is very lucrative for the Pharma companies.
Atomwise has the AtomNet platform which is the first of its kind, deep learning neural network for structure-based drug design and discovery. In one of its success stories, Atomwise discovered a drug candidate which may significantly reduce Ebola infectivity. It helped accelerated the search from thousands of approved medicines. This analysis, which typically would have taken months or years, was completed in less than one day.
3Scan’s aims to turn tissue biology and histopathology into a data science in order to translate these discoveries into new diagnostic and therapeutic solutions. They take a tissue sample analysis in one day that it would take a pathologist to do in one year using traditional methods.
Pharma giants like GlaxoSmithKline, Merck & Co, Johnson & Johnson and Sanofi are investing into AI technology systems to predict how molecules will behave and how likely they are to make a useful drug, thereby saving time and money on unnecessary tests.
Virtual Nurses / Robotic Companions for Patients and Elderly
Hello Nurses, are you ready for AI? — is an interesting article which talks about the AI impact on Nurses’ jobs. Taking temperature, blood pressure readings and any other repetitive tasks are fast being replaced by machines and it will be more accurate and efficient. Robotic companions and caregivers for the elderly is another growing area of AI which is finding lot of application especially in case of people with chronic ailments and diseases.
Remote Patient Monitoring using AI Virtual Assistants and Chatbots
Using Healthcare AI bots, patients can chat with an app about symptoms and self-care. Biometric information using wearable technology can be captured and accessed by the patient as well as the doctor and AI apps can be designed for particular outpatient programs without needing to schedule a clinic visit. Remote patient monitoring is turned out to be a boon especially for patients with chronic ailments needing extended duration of care that can now be availed from the comforts of their homes.
Preventive Diagnostics using Predictive Analytics
In Predictive Analysis, algorithms use lab reports, past medical history, demographic, behavioral data to identify health risks and issues.
Careskore uses its Zeus algorithm for predictions and risk profiling of the patient population and decide how likely a patient will be readmitted to a hospital. This analysis is used to improve quality of care.
Researchers are already applying machine learning technology to help predict the onset of seizures for persons with epilepsy. Remote monitoring of heart rate of outpatients and related predictions can also alert on the onset of a cardiac incident and intervention can be provided before chronic conditions become a medical emergency.
AI in Oncology and Cardiology
AI in cancer diagnostics is a getting a lot of attention. Early cancer detection, predictions on cancer risk as well as AI tools augmenting the doctor’s approach to charting treatment plans are areas that are currently being worked on.
Oncora Medical has a data analytics platform that can help doctors design radiation treatment plans for patients. They use predictive analytics to create personalized treatment.
In cardiology, — ECG analysis platform, robotic echo, remote identification of arrhythmia (irregular heart rhythm), predictions for cardiac risk profiling are some areas on the radar for AI work.
Cardiologs has ECG analysis platform to augment cardiologists by helping them speed workflow, providing highest diagnostic yield for the least physician effort. There is also the integration of Ultrasound, Artificial Intelligence and Robotic Echo which is a promising technology to watch. In Robotic echo, echocardiography researchers have successfully performed ultrasound exams using a robotic arm controlled via the internet on patients located at remote locations. AI in Cardiology is carving new pathways to address cardiac ailments and is very promising.
Conclusion
We are on an AI curve of exponential growth. It is most visible in the Healthcare industry and promises immense benefits in the near future. The environment around the physician/doctor is changing fast — the diagnosis and treatment is augmented and accelerated with AI based tools, new breakthroughs in life threatening diseases like cancer and cardiac, predictive risk profiling, robotic tools enabling remote monitoring and treatment, elderly care, virtual nursing are all taking new shape in the AI paradigm. If you look into the crystal ball — it beyond doubt shows the nursing homes, hospitals, and over-all medical ecosystem in the next few years presenting a drastically different picture than what we see today.
References
What AI-enhanced health care could look like in 5 years
Top Artificial Intelligence Companies in Healthcare to Keep an Eye On — The Medical Futurist
How Artificial Intelligence Will Change Medical Imaging
About the Author
Deepa is a founding member of Humans For AI, a non-profit focused on building a more diverse workforce for the future leveraging AI technologies. Learn more about us and join us as we embark on this journey to make a difference!
