Generative AI: Increasing efficiency of drug design

16

October

2023

No ratings yet.

Since the pandemic, the pharmaceutical industry has been redefining and reinventing its operations and implementing more digital innovations. Generative artificial intelligence is one of the digital technologies promising to accelerate and improve the drug discovery process. The drug discovery process is extensively complicated and requires significant investments of time and money. Realizing new drugs take, on average, between 12 to 18 years, costing approximately $2.6 billion. Eventually, only 10% of drugs make it to clinical trials. Several companies are extensively researching the possibilities of integrating generative AI to improve the efficiency of their drug discovery processes (GlobalData Healthcare, 2023). My interest in (holistic) health, medical innovations, and breakthroughs stems from my own medical journey, searching for remedies for my chronic condition. The possibilities shown by the technique of generative AI to support the development of new drug discovery genuinely intrigue me. Seeing through the years how innovations widen the possibilities in the medical landscape is exciting to me.

Companies train their artificial intelligence to inspect vast and complex chemical and biological data sets. Subsequently, generative models process all this data to locate new targets for treating diseases and create new molecular structures with suitable properties. The input of scientists is to look for specific molecules with particular characteristics to transform these into new drugs (Nouri, 2023).

One of the companies to use generative AI to discover new drugs is Insilico Medicine.  Insilico Medicine uses its generative AI platform, Pharma.AI, in each step during the drug discovery process. Traditional discovery of a drug for idiopathic pulmonary fibrosis would have cost over $400 million over six years. However, with the use of their generative models, the cost was $40 million, and the first phase of clinical trials began after 2.5 years. The generative approach to designing drugs thus increases time and cost efficiency (Yao, 2023).

Insilico Medicine even received IND approval from the U.S. Food and Drug Administration (FDA) to start their drug in the clinical validation stage(Insilico Medicine receives IND approval for novel AI-designed USP1 inhibitor for cancer, 2023). Besides, they have several other AI-designed drugs in their pipeline. Their lead drug for progressive idiopathic pulmonary disease is a breakthrough for entirely generated AI drugs since it is in phase II patient trials (Nouri, 2023).

Even though generative-designed drugs have a promising outlook, they also have barriers. Regulatory and ethical considerations may limit the design of drugs with the use of generative AI. Additionally, datasets must be high quality and large enough for machine learning (GlobalData Healthcare, 2023).

I believe using generative AI to design drugs will mature since it can shorten the discovery process of drugs and increase cost-efficiency. What do you think the future of healthcare will look like now that drugs can be created more efficiently with generative AI? Do you regard generative AI as a prescription for success in the future of healthcare?

References

GlobalData Healthcare. (2023, 3 augustus). Generative AI has the potential to revolutionise drug discovery. Pharmaceutical Technology. https://www.pharmaceutical-technology.com/comment/generative-ai-revolutionise-drug-discovery/?cf-view&cf-closed

Insilico Medicine receives IND approval for novel AI-designed USP1 inhibitor for cancer. (2023, 25 mei). EurekAlert! https://www.eurekalert.org/news-releases/990417

Nouri, S. (2023, 5 september). Generative AI drugs are coming. Forbes. https://www.forbes.com/sites/forbestechcouncil/2023/09/05/generative-ai-drugs-are-coming/

Yao, R. (2023, 27 juni). Insilico Medicine uses generative AI to accelerate drug discovery | NVIDIA blog. NVIDIA Blog. https://blogs.nvidia.com/blog/2023/06/27/insilico-medicine-uses-generative-ai-to-accelerate-drug-discovery/

Please rate this

My vision of the medical treatment industry in Germany in the year 2050 – Utopia or rather a dystopia?

9

October

2019

5/5 (2) The problems in the German medical treatment industry are multifarious (Müller, 2018). Besides having unnecessary and badly executed medical treatments, medicine is costly (Tautz, 2018) and people from the countryside are suffering from a decreasing number of medical care due to the rural exodus of many doctors (Kölsch, 2018). For example, Mecklenburg-Vorpommern has a need for doctoral replacement of around 25% because of the previously mentioned reasons (Korzilius, 2008). Nonetheless, not only the countryside is suffering from an undersupply of doctors as 52 000 doctors are expected to retire in Germany until 2020 (DAZ, 2010), but also hospitals are missing 80 000 caregivers currently (Heine, 2018). Furthermore, the absence of IT-networks or standards for data transfer (Banse, 2018) are fundamental reasons for inefficiencies and a poor allocation of resources in the Germany medical treatment industry (SVZ, 2017).

However, change drivers such as the technological development, digitization and new customer needs could potentially enable an enhanced medical treatment in the future (Gerst, 2015). First, the E-Health trend impacts the interaction between patient and service provider and simplifies the self-management of the patient via (mobile) health applications (Wicks, 2014). Secondly, technological developments such as the advancements in big data analysis, self-learning AI deep learning algorithms or the digitization in general allow an improvement in the analysis of patient data, better forecasts, prevention of upcoming illnesses and a rectified interconnectivity between the stakeholders in the medical treatment field (Ehneß, 2018). Additionally, the technological development also offers advancements on the hardware side. For example, hyperloop systems or drones could potentially allow a different medical treatment infrastructure (Rosser, 2018). Last but not least, biotechnical developments in genetic manipulation (Miller, 2018) or in reproduction of organs could facilitate a lifesaving opportunity for patients (Wallace, 2018).

In the following part I will elaborate on my vision for the medical treatment in Germany in the year 2050. In order to empower a vital discussion, I would be keen on knowing if you can identify with my vision of medical treatment in the year 2050. Ask yourself, if ethical aspects such as morality or freedom are considered.

1. Home (-station) treatment
The HomeStation is an interactive diagnostic and robotic system for home use. It can take over general medical tasks, replaces or supports nursing staff and thus guarantees 24/7 medical care. Part of that home treatment is the use of wearables, for example electronic medical tattoos or sensors, which are on the one hand able to measure data regarding blood sugar, respiratory rate etc. (Kraft, 2019) and are on the other hand able to transmit that data to the relevant device or doctor. The role of the doctor will be taken by a robot (Yasa, 2018) who will consult the patient based on 24/7 tracked data. In addition, the robot performs minor medical treatments such as blood sampling or vaccinations. Finally, a 3D-printers ensures an immediate supply of medication, prevents drug abuse and provides better drug treatment through networking with other systems (Soleil, 2019).

2. Stationary care
Stationary care includes supra regional hubs, local hospitals and hubs of expertise for special medical fields. These are connected via drones and an underground network of hyperloops to ensure a fast and efficient treatment of every patient, independent of the location of the patient. Treatment at the surgery will be performed by surgical robots (Crawford, 2016), which are more precise, faster and risk-free. Therefore, badly executed medical treatments can be avoided. Additionally, due to the development in biotech, new organs can be delivered on demand and personalized (Pollack, 2018). A further benefit of the advancements in biotech is the prenatal and postnatal repair of severe genetic defects through genetic manipulation (Sakuma and Yamamoto, 2018).

By using this vision as a guiding principle, the medical treatment industry improves in terms of interconnectivity, flexibility, resource allocation, quality, costs and equality of treatment. Nevertheless, the risks are ubiquitous as there are side and ethical effects of genetic manipulation, as well as a reduction of human individuality by using robots. Therefore, the question arises if humankind should detach itself from its natural state and “design” people by reproducing organs? I am really looking forward hearing your opinion on this very relevant topic.

References:
Banse, P. (2018) Digitalisierung der Medizin – Das deutsche Gesundheitswesen ist zu wenig vernetzt. Available at: https://www.deutschlandfunkkultur.de/digitalisierung-der-medizin-das-deutsche-gesundheitswesen.976.de.html?dram:article_id=413494 (Accessed: 5 October 2019).

Crawford, M. (2016) Top 6 Robotic Applications in Medicine – ASME. Available at: https://www.asme.org/topics-resources/content/top-6-robotic-applications-in-medicine (Accessed: 5 October 2019).

DAZ (2010) Neue Studie zum Ärztemangel: Knapp 52.000 Ärzte gehen bis 2020 in Ruhestand. Available at: https://www.deutsche-apotheker-zeitung.de/news/artikel/2010/09/03/knapp-52-000-aerzte-gehen-bis-2020-in-ruhestand (Accessed: 5 October 2019).

Ehneß, S. (2019) Wie sieht die Medizin der Zukunft aus? Available at: https://www.healthcare-computing.de/wie-sieht-die-medizin-der-zukunft-aus-a-833099/ (Accessed: 5 October 2019).

Gerst, T. (2015) Zukunft der Medizin: Trendstudie will den Weg weisen. Available at: https://www.aerzteblatt.de/archiv/171346/Zukunft-der-Medizin-Trendstudie-will-den-Weg-weisen (Accessed: 5 October 2019).

Heine, H. (2018) Personalmangel in Krankenhäusern: 35,7 Millionen Überstunden – Politik – Tagesspiegel. Available at: https://www.tagesspiegel.de/politik/personalmangel-in-krankenhaeusern-35-7-millionen-ueberstunden/22706004.html (Accessed: 5 October 2019).

Kölsch, T. (2018) Medizinischer Nachwuchs: Landflucht und Landarzt-Mangel. Available at: https://www.general-anzeiger-bonn.de/ratgeber/fit-und-gesund/landflucht-und-landarzt-mangel_aid-43810019 (Accessed: 5 October 2019).

Korzilius, H. (2008) Hausärztemangel in Deutschland: Die große Landflucht. Available at: https://www.aerzteblatt.de/archiv/59015/Hausaerztemangel-in-Deutschland-Die-grosse-Landflucht (Accessed: 5 October 2019).

Kraft, D. (2019) 12 innovations that will revolutionize the future of medicine, National Geographic magazine. Available at: https://www.nationalgeographic.com/magazine/2019/01/12-innovations-technology-revolutionize-future-medicine (Accessed: 5 October 2019).

Miller, J. (2018) The Future of Medicine. Available at: https://hms.harvard.edu/news/future-medicine (Accessed: 5 October 2019).

Müller, T. (2018) Gesundheitssystem Deutschland: Trotz hoher Gesundheitsausgaben – bei der Lebenserwartung hinken wir hinterher. Available at: https://www.aerztezeitung.de/medizin/krankheiten/herzkreislauf/article/976013/deutschland-hohe-gesundheitsausgaben-und-geringe-lebenserwartung.html (Accessed: 5 October 2019).

Rosser, J. C. et al. (2018) ‘Surgical and Medical Applications of Drones: A Comprehensive Review’, JSLS : Journal of the Society of Laparoendoscopic Surgeons. doi: 10.4293/JSLS.2018.00018.

Sakuma, T. and Yamamoto, T. (2018) ‘Genome editing for dissecting and curing human genetic diseases’, Journal of Human Genetics, 63(2), p. 105. doi: 10.1038/s10038-017-0380-0.

Soleil, V. (2019) 10 Possible Medical Treatments of the Future. Life Advancer. Available at: https://www.lifeadvancer.com/possible-future-medical-treatments/ (Accessed: 5 October 2019).

SVZ (2017) ‘Ländervergleich: Medizinische Versorgung: Gut ausgestattet, aber ineffizient. Available at: https://www.svz.de/deutschland-welt/politik/medizinische-versorgung-gut-ausgestattet-aber-ineffizient-id18296516.html (Accessed: 5 October 2019).

Szent-Ivanyi, T. (2014) Unnötige Todesfälle in deutschen Kliniken. Available at: https://www.fr.de/ratgeber/gesundheit/unnoetige-todesfaelle-deutschen-kliniken-11233271.html (Accessed: 5 October 2019).

Tautz, D. (2018) Gesundheitssystem: Hohe Kosten, trotzdem Mittelmaß. Available at: https://www.zeit.de/wissen/gesundheit/2018-03/gesundheitssystem-deutschland-bruttoinlandsprodukt-lebenserwartung (Accessed: 5 October 2019).

Wallace, L. (2018) Reproductive tech will let future humans inhabit the body they truly want, Clinical Endocrinology. doi: 10.1111/j.1365-2265.2009.03625.x.

Wicks, P. et al. (2014) ‘Innovations in e-health’, Quality of Life Research. doi: 10.1007/s11136-013-0458-x.

Yasa, D. (2018) Why robots could soon replace our doctors. Available at: https://www.dailytelegraph.com.au/lifestyle/health/body-soul-daily/why-robots-could-soon-replace-our-doctors/news-story/9c33db2f25e0fff6184603b38cdc641f (Accessed: 5 October 2019).

Please rate this

Can machines replace doctors anytime soon?

11

October

2017

No ratings yet. With each day some new developments in data science and machine learning are being published. The researchers constantly invent new ways to improve the performance of their predictive algorithms, thus improving the accuracy of the computers’ predictions on any given subject. One of such subjects, which should be of particular interest to us – humans – is medicine.

With the constant increase in the computation power and exponentially growing amounts of data, the researchers could potentially save a lot of lives with the use of advanced data science solutions. For example, research presented in the journal of the American Academy of Neurology states that the potential for artificial intelligence in precision medicine is significant. IBM Watson, a question answering computer system, provided a report of actionable insights within 10 minutes, in comparison to 160 hours of human analysis normally necessary to reach analogical conclusions (Monegain, 2017). This can often be of crucial difference when dealing with illnesses such as malign cancers, as some of them have a median survival of less than a few months following diagnosis.

Aside from genome analysis, there are countless areas in medicine where machine learning and artificial intelligence solutions can mean a difference between life and death (or very serious health complications). Currently, researchers around the world are trying to incorporate data science applications in detection of diseases such as autism, Parkinson’s, Alzheimer’s to name just a few! All of these sicknesses have one thing in common: early detection always improves the chances for a recovery or prevents further complications. And in this very sense, the AI can help people. Not only it is able to recognize the symptoms of a disease faster that a human can, but also it is able to do so at a larger scale, providing necessary diagnosis to the people in need. For instance, Google is launching an experiment to use machine learning to discover a diabetes-related eye disease in India, where the number of people with diabetes is around 70 million (with approximately 400 million people worldwide) and not a large percentage of them would normally receive a proper diagnosis in time (Simonite, 2017).

Another important factor, also connected to the accessibility, is the fact that a lot of people must postpone visits to the doctors because they cannot afford a private visit, while the wait time for a public one can even be around a year. And with the use of computers the time needed for an accurate diagnosis, as well as the number of specialists to be seen before obtaining it, is drastically reduced (Molteni, 2017).

Of course there are disadvantages of using AI in medicine. Firstly, there is a rising fear among the doctors that they will lose their jobs and be replaced by machines. According to an article in the New England Journal of Medicine, radiology and pathology are primarily susceptible to the power of AI, due to the fact that these jobs are based on pattern matching and machines can perform such tasks with surprising accuracy and speed (Asay, 2017). However, AI should mostly complement the work of doctors, enable them to perform their job more efficiently, thus help more people. Secondly, there are many issues regarding the data. Some people are not always aware that by signing a particular form they enable a company (or companies) to use their private data for research purpose (Molteni, 2017). What is more, for the algorithms to have a high accuracy, they need to have a lot of data at their disposal, which is not always feasible. For example, in the process of detecting the case of autism among kids, a lot of data needs to be gathered from MRI scans. This is not only time-consuming, but also expensive procedure. And it would not be realistic to have every child scanned in order to have a sufficient dataset (Vlasits, 2017).

Summing up, the rapid development of artificial intelligence and big data promises a future in which computers will be able to assist the doctors in providing quick and accurate diagnosis, thus saving human lives. However, one cannot forget that the algorithms should assist the doctors, instead of taking over their jobs. There will always be a fear that a machine provides an inaccurate prognosis, which can either scare the patient or make him calm, when he or she can be seriously sick. That is why human input is always valuable, to evaluate the machine’s diagnosis and verify its correctness. In other words, we should be very optimistic about the possibilities offered by artificial intelligence, but at the same time should not expect a bunch of robots treating patients in the hospital anytime soon.

References:

Asay M. (2017, January). Why AI is about to make some of the highest-paid doctors obsolete. Tech Republic, retrieved from: http://www.techrepublic.com/article/why-ai-is-about-to-make-some-of-the-highest-paid-doctors-obsolete/

Molteni, M. (2017, August). Want a diagnosis tomorrow, not next year? Turn to AI. Wired, retrieved from: https://www-wired-com.eur.idm.oclc.org/story/ai-that-will-crowdsource-your-next-diagnosis/

Molteni, M. (2017, September). 23andme is digging through your data for a Parkinson’s cure. Wired, retrieved from: https://www-wired-com.eur.idm.oclc.org/story/23andme-is-digging-through-your-data-for-a-parkinsons-cure/

Monegain, B. (2017, July). AI can speed up precision medicine, New York Genome Center-IBM Watson study shows. Healthcareitnews, retrieved from: http://www.healthcareitnews.com/news/ai-can-speed-precision-medicine-new-york-genome-center-ibm-watson-study-shows

Simonite, T. (2017, June). Google’s AI Eye Doctor Gets Ready to Go to Work in India. Wired, retrieved from: https://www.wired.com/2017/06/googles-ai-eye-doctor-gets-ready-go-work-india/

Vlasits, A. (2017, June). AI could target autism before it even emerges – but it’s no cure-all. Wired, retrieved from: https://www-wired-com.eur.idm.oclc.org/story/ai-could-target-autism-before-it-even-emerges-but-its-no-cure-all/

Please rate this

Future of Medicine: Will robots replace doctors?

21

September

2016

No ratings yet. In 2011, IBM computer Watson won first prize in the American TV quiz show “Jeopardy.” Watson spent three seconds finding the right answer to every single question he was asked, and in those three seconds the computer went through hundreds of processes all at once. Watson can not only understand the questions, he can also quickly find the correct answers and then proceed to communicate these. When winning “Jeopardy”, Watson had to understand and analyze references and metaphors, such as “feeling blue.” His victory in “Jeopardy” has great entertainment value and of course it became a hit on YouTube (link to video). But behind the “brains” lies three years of intense efforts by 20 researchers, who developed a technology called DeepQA. The aim was to create a new generation of technology that could find answers in unstructured data, and in a much more efficient way than previous technology has managed. The revolutionary aspect of Watson is that it can understand the world much the same way that humans do, through senses, learning and experience.

How can we take advantage of this in the health care industry?

Health is the most exciting area that Watson is able to revolutionize and earlier this year he helped solving a medical mystery.  Doctors at a hospital in Tokyo had tried for a really long time to find the right treatment for a woman with leukemia, but every effort to combat the disease failed. When IBM’s genius computer Watson took over, he only used ten minutes to study the patients’ medical records and cross checked this information with over 20 million oncological reports. Watson concluded that the woman did not have the type of leukemia that doctors initially thought, but rather a very rare variant which required a different type of treatment than the one she had been receiving, according to Siliconangle.

Understands “common” language

Computers have long been of assistance in the health sector, but one of the biggest challenges until now has been that the machines do not understand natural language. Due to the fact that Watson understands this and can respond to it, it is no longer necessary to translate research articles, treatment guidelines, patient records and hospital records, textbooks, notes and emails to structured computer language. A doctor can ask Watson a question and describe the patient’s symptoms and other relevant factors. Watson will then analyze the information from the doctor and combine it with the latest research results and examine all available sources. When Watson is done with the analysis, a list will come up with potential diagnoses, together with an estimate that shows the validity of each diagnosis.

This new technology may revolutionize the health care industry as we know it today. However, will it ever become so good that it can replace the human medical professions?

 

Sources:

http://researcher.watson.ibm.com/researcher/view_group.php?id=2099

http://siliconangle.com/blog/2016/08/05/watson-correctly-diagnoses-woman-after-doctors-were-stumped/

http://www.ibm.com/smarterplanet/us/en/ibmwatson/health/

 

 

Please rate this