Artificial Intelligence #AI – the new hope for Pharma R&D - By Manishree Bhattacharya @ManishreeBhatt1

Pretty much every article starts with the challenges that pharmaceutical industry across the globe is facing. It is a difficult industry and everybody acknowledges that, considering the time to develop an original drug (10-15 years), the costs involved (last time I checked it was USD 2-3 billion), the high attrition rates of drug candidates (1 out of 5,000 or 10,000 leads make way for FDA approval), the tough regulatory environment which is varied across countries and geographies, and the rising pressures on pricing (pricing advantage for truly outcome-driven therapeutics). All of these, with the looming patent expiry, the imminent entry of generics, and the tantalizing RoIs, make it even more difficult.
Pic Credit:
Well, technology is here to rescue. It will be a grave injustice to talk about technology implementation in pharma R&D in just one article, and because AI and ML are the current buzzwords, I thought, maybe, we could specifically discuss the role Artificial Intelligence plays in streamlining and improving success rates of pharma R&D. Before we go any further, let us have a quick look at the drug discovery process and the typical timelines associated.
More often than not, it takes more than a decade (sometimes > 15 years) for a new drug to enter the market. One should not forget if technologies and in-silico modeling are not effectively used in early stages of drug discovery, drug failure at a later stage is a significant waste of time and money. Now, pharma companies have already been using computational tools to conduct ADMET predictions and in-silico modeling, so what new has Artificial Intelligence to offer?
To answer this question, important will be to look at some of the AI initiatives and activities of key pharmaceutical companies. You may find initiatives where pharma companies are scouting for AI innovation via their open innovation program, so if you are a tech start-up, think about it. 
The list is not exhaustive and is only for the purpose of illustration, but one can clearly see that leading pharma companies have already dipped their toes in Artificial Intelligence-mediated drug discovery and development.
Going back to the drug discovery process diagram, let us superimpose the AI-applications across the process.
Artificial Intelligence and Machine Learning algorithms, not only simplify the existing tasks/processes, it saves time, while adding significant value. Let us understand this better.
BioXcel Corporation, a biopharmaceutical company is working on integrating big data and AI into drug discovery process. One such product is EvolverAI which uses AI algorithms for drug discovery to find the best therapy, thus reducing drug failure. Evolver AI uses big data to screen through huge volumes of structured and unstructured data related to genes, proteins, disease pathways, targets, symptoms etc. within the field of Neuroscience. This is followed by creation of meta-data, which contains network-maps, linking pathophysiology of diseases with drugs. These meta-data are then fed into a decision matrix, which compares all the drugs. Using AI algorithms and human intelligence, several hypotheses are built based on the known linkages, of which the best hypothesis gets selected for future experiments and clinical trials, significantly reducing both time and risk of drug failures.
Exscientia, which has partnerships with both GSK and Sanofi, uses AI to learn best practices from drug discovery data, and helps researchers generate drug candidates in much lesser time.
The beauty of such algorithms is in their ability to go though various data-sets – from medical records, publications, clinical trial data, available data on disease pathways and drug-disease correlation to derive meaningful analysis that can help researchers in decision making. Similarly, screening through EHR, current and past clinical trials, and available publications on epidemiology, can help in site selection for clinical trials. Applications are many, and this is still an evolving area, with many developments happening in the stealth mode.
Many diseases such as cancers and neurodegenerative disorders, having high mortality and morbidity, have been debilitating for people across the globe, with pharma companies pouring in significant amount of money, yet with not-so-significant success rates.
Take Alzheimer’s for example, after three large clinical trials for solanezumab, which was called the breakthrough drug candidate, it hasn’t been able to display significant change in patient’s conditions as compared to a placebo. Solanezumab is an anti-Aβ mAbs, that targets Amyloid beta, which is one of the main components of the amyloid plaques found in the brains of Alzheimer’s patients. Does that mean researchers are looking in the wrong place? Is there a need to rethink the disease pathway and establish newer biomarkers to be targeted? THESE are the areas where AI and ML can help.
Similarly, increasingly the drug research community is realizing that not all drugs work on all patients. AI algorithms can help stratify patient population to identify what causes drug efficacy in some patients while nothing really in others - could be due to an aberration, mutation, any specific biomarker – and AI can help recognize the same in the most efficient manner possible.
And that is not all, there is a lot more - we are just scratching the surface, but one cannot deny that these technologies are the next hope for pharmaceutical companies. What do you think?
Source: Nature, Wall Street Journal, MIT Technology Review, Forbes, PR NewsWire, Micar21, SlideGeeks, BenchSci, Xconomy, MedCityNews, Company Websites
[content title="About Manishree Bhattacharya" icon="fa-heart"]
Manishree Bhattacharya
Independent consulting – strategic research, industry analysis, healthtech evangelist, digital thought leadership (ex-NASSCOM, ex-Evalueserve)
Has over 8 years of experience in strategic research across healthcare, life sciences, software products, and start-ups, and has extensively worked with Indian and Global clients, helping in market analysis, digital evangelizing, start-up collaboration, competitive intelligence, and decision making
[content title="Latest Articles"]
[/content] [/tab]

Read more »

Artificial Intelligence #AI Could Add $957 Billion to Indian Economy, According to New Research by @AccentureIndia

In a recently published report by Accenture, they have highlighted the need for india to invest in AI, we bring you the excerpts of the report. (The following content is sourced from the Accenture report).

Artificial intelligence (AI) has reached a tipping point. The combination of the technology, data and talent that make intelligent systems possible has reached critical mass, driving extraordinary growth in AI investment. Across the world, G20 countries have been building up their AI capabilities. The power of AI starts with people and intelligent technologies working together within and across company boundaries to create better outcomes for customers and society. But India is not fully prepared to seize the enormous opportunities that AI presents. Even with a tech-savvy talent pool, renowned universities, healthy levels of entrepreneurship and strong corporations, the country lags on key indicators of AI development. Much work remains. 

Read more »

#AI in Healthcare by @deveshrajadhyax

There are some subjects that invoke sharp and contrasting emotions in the society. In present day India, the GST tops the list of such things that are considered boon by some and curse by others. New technology usually does this to people. The steam engine, the telephone and the computer all have been greeted both as the savior and the nemesis of the mankind. 

Read more »

Four ways in which #AI can help humankind @deveshrajadhyax

Artificial Intelligence is receiving more than its fair share of public attention. On one side there are promises of miracles, while on the other side there are warnings of doomsday. What is probably missing is a simple listing of clear benefits. This is article is an attempt to create such a list.

Read more »

Understanding the Medical Diagnosis processes, to build an AI based solution by @msharmas

Human Intelligence

is The ability to adapt one's behavior to fit new circumstances.

In Psychology, human intelligence is not regarded as a single ability or cognitive process but rather as an “array” of separate components. Research in building AI systems has focused on the following components of intelligence: [1]
  • learning,
  • reasoning,
  • problem-solving,
  • perception, and
  • Language-understanding
These components of human intelligence are also utilized during diagnosing a patient and defining the treatment plan and protocol for the patient.

Read more »

Is superintelligence the inevitable next step in evolution? by Dr. Roshini Beenukumar, @roshiniBR

Experts predict that by 2050, there is a 50% probability that AIs which will match the intelligence of an average adult human. It is not a long wait, isn’t it?

Last week, my husband and I finally sat down to plan our summer vacation. After spending two full hours on 10 different tour websites, we ended up being completely overwhelmed by the options; all-inclusive vs half-board, city vs seaside or sea vs land. 

Read more »

The Current Status of 8 Future Technologies on Healthcare by @msharmas

It's mid-2016, and here is a look at the current status of 8 Future Technologies that might be having a significant impact on Healthcare

Most if not all these technologies will make an impact on Healthcare, and hence it is important to understand the various scenarios and the stories detailing how the experts from across the world are incorporating these technologies in healthcare

Read more »

House MD vs Doctor #AI- Who will turn out to be the better by @RoshiniBR

House MD vs Doctor AI- Who will turn out to be the better diagnostician?

Do a google search for “all-inclusive beach holiday” and all you see for the next weeks are advertisements for all-inclusive holidays following the virtual you. Google knows you better than your family or friends- creepy but true! Whether or not we realize it, we have made the decision to donate our data to the virtual world.

Can we put our data to better use – to improve healthcare on a scale unimaginable a decade or two ago?

Read more »

Benefits of an AI-Based Patient Appointments service for Hospitals by @msharmas

One of the areas where AI can be implemented in the Hospital with high volume of transactions, is the Appointments Scheduling of Patients. On any given day, there are a finite number of slots available for a doctor, e.g. 10 min or 30 min slots, depending on whether its a first visit or a follow up visit. In most hospitals, Routine patients are scheduled in advance and some patients are scheduled based on an urgency, to the physician schedule.  [Denton et al - 8]
A typical workflow for booking an appointment can go like this:

1. Patient calls (or visits) the hospital, and speaks to the person at the reception, at a specific department
2. The person looks up the available time slots, that a doctor is free and available in the clinic
3. Consults with the Patient on the best time possible for her appointment and then schedules the appointment

Read more »

Zen Clinicals: An Activity & Workflow based solution (1 of 3)

Part 1 of 3:
Recently, during the Gartner Symposium, it was predicted that cognitve platforms would take over a lot of activities. Keeping this future at the back of our minds, its important for the EHRs of today to metamorphosize to the Cognitive Computing platforms of tomorrow, and fast.

With social media, predictive and bigdata analytics become more central to the discussions; the current EHR systems are woefully lagging behind in the ability to catchup with the technology of today. Newer EHRs or healthcare based platforms too are lagging behind in the adoption of these technologies. Most often have observed the similar functionality being rolled out on newer technology.

In most cases I have found the solutions to be non-workflow, i.e., non-BPM, enabled solutions that have the inability to adapt to a new healthcare scenario faced by the solution in the next implementation. With the advent of the Software as a service platforms and the proliferation of the cloud based services in other industries, healthcare solutions have generally gone for the traditional (non-BPM based) approach to delivering a solution. 

So, One needs to then ask, "What should the future of a patient health record (PHR) or the electronic health record look like?" 

Will we really see the advent of SMART solutions in healthcare, or are the changes to difficult too incorporate into the solution because that's how the solutions had been developed?

In this series I present some thoughts on what should the solutions of today morph into to meet the needs of the users and hopefully make it more user-friendly in the process.

And to develop the solution, Zen Clinicals (this is a fictitious name of a solution any resemblance is purely a coincidence), we need to answer a few questions upfront: 
  • who do you build it for? 
  • Who are the actors? 
  • And what do they need? 
  • What are their activities? 
  • What do they do daily?

“User Experience strategy lies at the intersection of UX design and business strategy. It is practice that, when done empirically, provides a much better chance of successful digital product than just crossing your fingers, designing some wireframes, and then writing a bunch of code”

UX Strategy by Jamie Levy (O’Reilly)

Read more »

Popular Posts