The Neuro Spark #13: Neuroscience news impacting health tech
Every week we select 5 articles highlighting what is going on in the health tech industry and research. You should definitely have a look at them.
”How Deep Learning and Neuroimaging Can Improve Diagnosis in Psychiatric Diseases” by Irina Sanchez via QMENTA Tech Blog
Irina is working as a Deep Learning Engineer at QMENTA. Recently, she used Deep Learning to classify different brain regions and link them to psychiatric diseases. The aim is to introduce a game changer, deep learning, when applied to the diagnosis of psychiatric diseases. If around 450 million people are estimated to suffer from psychiatric diseases in the world, diagnoses are still mostly based on symptoms and subjective information. Deep learning can help by obtaining biomarkers from brain images to understand these type of disorders better. “Those biomarkers can be used to improve the accuracy of the diagnosis, and consequently selection of a personalized treatment for each patient” says Irina. Eventually, deep learning could come as support for doctors dealing with mental illnesses.
This is a very interesting read from Elysium Health presenting a guide that delves into the world of clinical trials. Clinical trials are essential for drugs to be put on the market, but there is a very complicated process before that happens. The article also provides us with a well-rounded glossary about the key expressions used in the field of clinical trials. Going about all the different phases and types of clinical trials shows the interconnections existing between all of the companies involved in the health industry. But, there is also a timeline of clinical trials tracing the first clinical trial all the way to the Bible.
“On average, it takes around eight years and costs roughly $2.6 billion based out-of-pocket and time costs to bring a new drug to market,” says Mary Jo Lamberti, Ph.D., associate director of sponsored research at Tufts Center for the Study of Drug Development.
“Researchers design artificial intelligence system for predicting drug combinations’ side effects” via News Medical Life Sciences
More and more people are taking drugs. Most of the times there is no knowledge of the side effects caused by adding a new drug to the routine of the patient. The number of drugs on the US market is amazing and to test one drug combination means 5,000 experiments. There are too many combinations and to test all of them would cost a lot of time to researchers. Computer scientists created an AI-based system able to predict potential side effects from drug combinations. Deep Learning can identify patterns in how side effects arise based on how drugs target different proteins, predicting potential side effects from drug combinations. “It was surprising that protein interaction networks reveal so much about drug side effects,” said Jure Leskovec, who is a member of Stanford Bio-X, Stanford Neurosciences Institute and the Chan Zuckerberg Biohub.
“Alzheimer’s breakthrough: brain metals that may drive disease progression revealed” via AANS Neurosurgeon
Recently, researchers made a breakthrough discovery regarding the properties metals in the brain that are responsible for the progression of neuro-degenerative diseases. Dr. Joanna Collingwood — a researcher at Warwick’s School of Engineering — helped qualifying iron species linked to the formation of amyloid protein plaques in the brain. Amyloid plaques are linked with toxicity and cell and tissue death, which causes mental deterioration in patients suffering from Alzheimer’s. The discovery was that in these patients, magnetite was forming in the plaques amongst other chemically-reduced iron species. Magnetite is a magnetic iron oxide and very rarely found in the brain. This comes from an interaction between amyloid protein and iron. Researchers from the UK and the USA managed to show the processes happening in the brain.
“The Role of Integrated Data in Achieving the Right Diagnosis and Treatment” by Homer Pien via Healthcare Tech Outlook
Homer Pien, PhD is working as chief scientific officer for diagnosis and treatment at Philips Healthcare. In his opinion, integrating radiology, pathology, genomics & electronic health records is something that is bound to happen. The medical industry is more and more digitalized, hence the need to integrate the increasing data produced in the health system. Because it is essential to establish an accurate diagnosis for the patient, these new technologies — integration and AI for instance — are narrowing down the details of the disease, enabling physicians to get the adequate treatment for their patients. Making the link between data extracted from radiology, pathology and genomics is going to accelerate the diagnosis and decision-making process ahelp take better care of patients.
“An accurante and definitive diagnosis plays a key role in the patient journey, as it sets the stage for treatment”
Who are we?
QMENTA is a brain imaging and data analytics platform in the cloud aiming to accelerate the discovery and development of cures for neurological diseases.