Ethical dilemmas in the healthcare market are complex moral problems that develop in the industry as a result of interactions between the financial aspects of a market-driven system and healthcare as a necessary service. Conflicts between moral principles, competing interests, and societal ideals are often common. These situations frequently lead to conflicts between competing interests, ethical norms, and social values.

Patients, families, and healthcare workers make ethical and legal decisions on a daily basis. The context of these challenging choices could be patient autonomy, clinical procedures, hospital management, or other issues that develop in the healthcare sector. Making decisions on behalf of patients who are unable to do so or considering the right to an abortion are issues that call for a more thorough, careful response. When moral issues in healthcare are taken into consideration, the actions taken clearly distinguish between what is right and wrong. It is important to note that healthcare in the future will likely be impacted by many decisions made today.

It is also important to note that in some countries, not everyone has equal access to medical care. It is also possible that people with higher incomes or better insurance receive better care, leaving out the rest.

Making ethical choices about end-of-life care, such as whether to give terminally ill patients aggressive treatment or switch to palliative care, can be challenging. There is also a growing concern about overdiagnosis and overtreatment in healthcare, where patients may be given pointless tests, procedures, or medications.

In situations where there are limited resources, healthcare professionals frequently have to make morally challenging decisions about who gets treated. When there are more patients in need of organs than there are organ donors, for instance, it can be very challenging to make decisions about organ transplantation. Other issues include health resource allocation, research ethics, data security and privacy as well as healthcare pricing and profit.

In these situations, solving these ethical issues in the healthcare industry is crucial because it has a direct bearing on patient care standards and the morality of healthcare organisations and providers.

First and foremost, regulatory bodies must offer stakeholders in the healthcare industry clear ethical guidelines based on accepted ideas like beneficence, non-maleficence, autonomy, and justice that serve as a moral compass. These guidelines provide a foundation for ethical decision-making and actions across the industry.

Back to Management

In the search for faster and more effective medicines, the integration of artificial intelligence (AI) can positively impact millions of lives. AI is transforming the foundations of biotech, promising to alter precision medicine, clinical trials and bioprocess optimisation. AI-powered analysis, prediction engines and cutting-edge robotics are propelling biotechnology into a future of customisation, greater efficiency and accuracy.

Dr. Alex Zhavoronkov, founder and co-CEO of Insilico, a healthtech enterprise, highlights the role of AI in bringing us closer to a precision medicine future. “We use our end-to-end Pharma.AI platform to analyse massive quantities of data and discover optimal targets for diseases. We also use generative AI to design new drug-like molecules that are optimised across 30 parameters to be highly effective. Advances in the use of AI for disease detection and diagnosis, as well as identifying likely drug interactions, are also accelerating us toward a precision medicine future, where treatments are optimised, based on AI data analysis, for a particular patient.”

With regards to clinical trial optimisation, Dr. Zhavoronkov explains that 90 per cent of drugs in development fail during clinical trials — most of them in Phase II trials with patients. He says it is critical AI tools are used to enhance predictions and guide decision-making to improve outcomes and ensure better success rates.

“We developed our AI tool, inClinico, to address this need. inClinico relies on various data - including OMICs data, clinical trials, patents, publications, grants and 13,000 drugs from Phase I to launch. Using machine learning and AI, we built models based on data points specific to successfully launched and failed drugs, including clinical trial design, biological features of the disease, drug targets and patient criteria. We combined these models into our AI clinical trial prediction engine. For every evaluated Phase II, inClinico generates a probability of success for proceeding into Phase III,” Dr Zhavoronkov says.

Bioprocess optimisation involves the use of advanced technologies, including AI and robotics, to enhance the speed, efficiency and accuracy of biological processes such as cell culture, high throughput screening, and genomics analysis and prediction – using autonomous guided vehicles (AGVs) including waste transportation, operation and storage robots and dual-arm composite lab robots. This can improve outcomes in biotechnological applications. Dr. Zhavoronkov explains how combining AI with robotics allows greater levels of speed, efficiency, and accuracy.

“With robots running labs, there can be more experiments, performed faster and in parallel, generating high-quality data that strengthens the AI’s ability to make accurate hypotheses and validate those hypotheses. As the robots perform experiments and generate results, data is fed back into the AI system,” he says.

Dr. Zhavoronkov’s insights also highlight AI’s transformative impact on biomanufacturing efficiency when seeking sustainable and cost-effective bioproduction. The generative AI platform at Insilico Medicine, for example, optimises production processes by designing molecules perfectly tailored to specific targets, exceeding the capabilities of traditional medicinal chemistry. This optimisation across 30 properties ensures efficacy and safety while minimising waste and maximising efficiency by synthesising and testing fewer molecules. The integration of AI with robotics can accelerate drug discovery, streamlining processes and continuously improving data quality.

AI can also streamline processes, improve drug discovery, and identify novel targets for diseases in record time. Insilico Medicine recently used AI to develop a key drug to treat idiopathic pulmonary fibrosis, a chronic lung disease. The orally delivered drug INS018_055 is the first fully generative AI drug (the first with both an AI-discovered target and designed by AI) to reach Phase II trials with patients. “AI was used at every stage and the drug has set many milestones in terms of speed of development. The preclinical candidate was selected in February 2021, just 18 months after the project began. Nine months later the Company announced first-in-human for Phase 1 trials, about half the time the process would take with traditional drug discovery.”

The transformative effects of AI, data and analytics on drug discovery are set to be life-changing. Scientists will be able to automate previously manual tasks and generate new insights at unprecedented speed. From revolutionising precision medicine through AI-driven analysis and generative design of drug molecules to enhancing clinical trial predictions with AI tools, AI promises to significantly improve drug development outcomes.

The combination of AI and robotics further accelerates bioprocess optimisation, encouraging speed, efficiency, and accuracy in biological experiments. AI’s role in sustainable biomanufacturing and its ability to expedite drug discovery, as evidenced by the success of INS018_055, highlights the profound potential AI has to reshape the future of medicine. 

Back to Technology

The introduction and continuous evolution of cardiac computed tomography (Cardiac CT) in the last two decades represents one of the most fascinating recent innovations in clinical medicine. The high accuracy and reproducibility of this rather new method together with exceptional predictive power makes cardiac CT a game changer in modern cardiology.

After years of stenosis-driven treatment decisions in coronary artery disease (CAD), characterised by the search for coronary luminal narrowing exceeding 70 per cent, to subsequently revascularise this stenosis through coronary bypass graft or coronary stent placement, the chapter of ischemia-centred treatment approach was opened. It became obvious that the pure anatomical approach might not be able to paint the full picture for optimised therapy, whereas information about the situation at the myocardium becomes more important.

The dynamic interplay between anatomy and function has driven the development of more sophisticated techniques to diagnose and manage coronary artery disease (CAD). Initially, cardiac CT has proven to have a very high negative predictive value (NPV) for the diagnosis/rule out of CAD. In cases of good diagnostic image quality — nowadays achievable in almost all patients thanks to the high technical performance of modern CT scanners — and normal coronary arteries, a CAD is excluded with an NPV of 97 – 99 per cent. It can be simplified that a negative CT safely excludes CAD.

Even the presence and severity of coronary artery stenoses can be detected and assessed with very high diagnostic accuracy. In other words, the prerequisites for anatomical imaging have been fulfilled by this non-invasive imaging technique. As proof of this development, “coronary CTA is recommended as the initial test to diagnose CAD in symptomatic patients in whom obstructive CAD cannot be excluded by clinical assessment alone”, according to the recent guidelines of the European Society of Cardiology (ESC) for the diagnosis and management of chronic coronary syndromes.

However, in times of ischemia-driven treatment decisions, this morphology-only approach represented by the rule out/diagnosis of relevant stenosis alone is not sufficient anymore as mentioned above. Cardiac CT has shown to be able to non-invasively provide the biomarkers needed for such ischemia-driven patient management by several unique features.

The accuracy of coronary stenosis assessment can be further increased by adding CT-derived Fractional Flow Reserve (CT-FFR) providing functional information about the pressure gradient among a stenosis. Additionally, possible myocardial perfusion differences can be detected in stress as well as in rest by adding CT perfusion. Last but not least, the possibility to characterise coronary plaques and predict future events has been introduced recently.

The underlying principle of CT-FFR involves simulating blood flow within the coronary arteries and calculating the pressure drop across stenotic lesions. By combining anatomical information from CCTA with computational modelling, CT-FFR generates a virtual fractional flow reserve value, reflecting the severity of coronary obstruction and its impact on blood flow. This innovative approach offers a more comprehensive understanding of coronary physiology, bridging the gap between anatomy and function.

One of the primary advantages of CT-FFR is its non-invasive nature, eliminating the need for invasive coronary angiography or pressure wire measurements. This not only reduces patient discomfort but also minimises the associated risks and complications. Moreover, CT-FFR provides a more detailed assessment of coronary lesions, enabling clinicians to differentiate between haemodynamically significant and insignificant stenoses.

The integration of CT-FFR into clinical practice has been supported by robust research demonstrating its diagnostic accuracy and clinical utility. Multiple studies have shown that CT-FFR has a high sensitivity and specificity for identifying functionally significant coronary stenoses when compared to invasive fractional flow reserve measurements. This evidence has contributed to the growing acceptance of CT-FFR as a valuable tool in the diagnostic armamentarium for CAD.

Myocardial CT perfusion (CTP) provides detailed information about myocardial blood flow and perfusion, aiding in the identification and characterisation of ischemic heart disease. The primary objective of myocardial CTP is to evaluate the blood supply to the heart muscle during rest and stress conditions. By utilising contrast-enhanced computed tomography scans, this imaging modality can visualise the distribution of contrast material within the myocardium, highlighting areas with compromised blood flow. During stress imaging, typically induced through pharmacological agents, the test can reveal regions of the heart that may exhibit reduced perfusion, indicative of obstructed coronary arteries.

One of the significant advantages of myocardial CTP is its ability to provide a comprehensive assessment of both anatomy and function in a single examination. The information obtained from myocardial CTP is valuable for guiding treatment decisions. Furthermore, myocardial CTP has shown promise in risk stratification and prognostication for individuals with suspected or known CAD. The ability to assess myocardial perfusion adds an important dimension to the diagnostic process, assisting clinicians in tailoring management strategies to individual patient needs.

Recently, cardiac CT has shown an ability to not only predict but also improve the outcome of patients. The Scot-Heart trial, a landmark trial published in 2018, and its subsequent studies proved the relationship between plaque-related biomarkers, identified by cardiac CT, and events during follow-up. Additionally, this study could also show that the addition of cardiac CT to normal patient care has the potential to improve the outcome by direct reduction of events and cardiac-related deaths. Remarkably, this was not achieved by an increased number of coronary revascularisations, but by optimising the medical care of the patients based on CT-based risk estimation.

The story of cardiac CT is still ongoing with an exceptional direct impact on patient care and outcome. The chapter of ischemia-driven treatment decisions will be followed by the just opened one of outcome-driven decisions, facilitated by non-invasive cardiac CT. The potential of plaque visualisation together with anatomical information and myocardial assessment is not yet fully explored but will for sure change patients’ pathways in the future. 

References available on request.

Dr. Christian Loewe is the Chairman of the Division of Cardiovascular and Interventional Radiology, Department of Bioimaging and Image-Guided Therapy at the Medical University of Vienna, Austria. 

The human 7 Tesla (7T) MR systems have enabled high-resolution and contrast imaging and have been applied to investigate brain pathology as well as neuroscience questions, especially in terms of high spatial resolution. The 7T/3T ratio is 2.33, and iso-SNR voxel size of isotropic 1 mm at 3T is isotropic 0.75 mm at 7T when the signal-to-noise ratio (SNR) increases linear to the magnetic field strength. However, many studies suggest supra-linear increase. The relative SNR of the whole brain at 7T compared with 3T is more than three times higher.

For T1-weighted imaging, magnetisation-prepared rapid 2 gradient echoes (MP2RAGE) are mostly used with a resolution of around 0.7 mm isotropic at 7T as compared to MPRAGE with an isotropic resolution of around 1 mm at 3T. In MP2RAGE, an image pair of 2 different inversion times are acquired, and image inhomogeneity is cancelled. In addition, fitting the longitudinal signal recovery of the two images provides a quantitative T1 map. Due to its high resolution, quantitative analysis of the cortex can be conducted. In normal aging, cortical thinning is well-known. In addition, T1 shortening is observed in aged subjects compared to younger ones (Figure 1).

Figure 1. Changes by aging in (Left) cortical thickness and (right) T1 values of healthy subjects. Cortical thinning and R1 (=1/T1) increase can be observed in aged compared with young healthy lubject.

As for the image contrast, T2*- weighted imaging (T2*WI) at 7T shows high contrast to myelin and iron. To detect subtle iron deposition, susceptibility-weighted imaging (SWI) is frequently used at 3T, however, high contrast for iron due to shortening of the T2* value may not require SWI at 7T, and T2*WI provide anatomical details and pathophysiological changes. Such high contrast of 7T is exploited in functional MRI (fMRI), where blood oxygen-level dependent (BOLD) contrast that is reflected on T2*WI.

Using its high SNR, sub-millimeter isotropic scans have recently been conducted. Laminar fMRI of the cortex is one of applications. Layer 4 of the cortex receives input mainly from the thalamus, and superficial layers mainly receive feedback from other cortical areas, whereas deeper layers mainly send information to other areas including the caudate nucleus. Investigation of layer-wise activation and/or connectivity is expected to elucidate the pathophysiology of neuropsychiatric disorders. Functional connectivity for small brain regions, such as habenula, has also been reported. Although it is a tiny structure located medial to the thalamus, the habenula is related to schizophrenia, depression, and other disorders. A network analysis of such key structures is expected to be an imaging biomarker of disorders.

Investigation of the neurochemicals may also contribute to the elucidation of pathological conditions. MR spectroscopy (MRS) can separately measure neurochemicals using their specific patterns of chemical shift peaks. The amount of chemical shift increases to the increase of the static magnetic field, enabling better separation of neurochemical peaks. The mostly measured neurochemicals at 3T were n-acetyl-aspartate, choline, creatine, myo-inositol, and lactate in addition to glutamate (Glu), which is a representative excitatory neurotransmitter, but at 7T, the inhibitory neurotransmitter, gamma- Aminobutyric Acid (GABA) can also be measured without spectral editing, although macromolecules need to be appropriately removed. Neuroprotective chemicals, such as glutathione and taurine, can also be measured. Their profiles are investigated to explore neurochemical biomarkers. In addition, dynamic changes of Glu during tasks can be observed (Figure 2), which is not very practical when spectral editing is used.

Figure 2. Changes of glutamate concentrations. An increase is observed during the finger tapping task (Tap) for 2.5 minutes compared to Rest conditions for 2.5 minutes.

It should be noted that the human 7T-MRI can also be used to investigate specimens. Mesoscopic information can be acquired using the same pulse sequences when dedicated coils are used to increase SNR. However, there exists several difficulties in 7T, such as increased inhomogeneity of both static magnetic field (B0) and radiofrequency transmit field (B1+) and increased specific absorption rate (SAR), resulting in larger signal dropout, signal inhomogeneity and lower contrast compared with 3 Tesla (3T). Aside from B0, B1+ inhomogeneity can be improved using dielectric pads or parallel RF transmission. SAR increases quadratic to the static magnetic field strength, and this is highly related to the safety. Considering these issues, 7T has already been accepted for clinical use for more than 5 years and will bring new insights into clinical practice. 

References available on request.

Tomohisa Okada, MD, PhD, is the part of the Human Brain Research Center, Graduate School of Medicine at Kyoto University in Japan.

The World Health Organization (WHO) reports that over two-thirds of the global population lacks access to radiology services. Emerging markets such as island nations and 14 African nations face critical shortages, where limited access to hospitals, advanced imaging equipment, and medical professionals impacts millions in need of radiological diagnosis and treatment. Even countries with robust healthcare systems, such as the US and Australia, face disparities in access between major cities and rural areas.

The rapid pace of innovation in radiology holds the promise of improving diagnostic accuracy, reducing costs, and enhancing accessibility in healthcare. This transformative evolution in imaging technology has the potential to address the talent shortage while providing clinicians with more precise and timely diagnostic information to improve patient outcomes and overall healthcare efficiency. The year 2023 paved the way for several key breakthroughs that cement wider adoption of several new radiology technologies.

Enhancing mobile medical imaging

The post-pandemic emergence of mobile medical imaging technology, image sharing, and storage has made it easier than ever to capture and share patient information such as x-ray, CT scans and MRIs with practitioners, while remaining HIPAA compliant and protecting patient privacy. This trend is expected to pick up pace as mobile medical imaging technologies continue to enable clinicians to deliver swift and cost-effective diagnostic imaging services to patients in remote or underserved areas. Notably, mobile computed tomography (CT) and mobile magnetic resonance (MR) imaging stand out as promising technologies, diagnosing and treating various medical conditions in diverse clinical settings.

These mobile units retain all the capabilities of their stationary counterparts but with the added advantage of portability. This allows physicians to take the equipment to the patient, saving time and reducing costs associated with patient transfers to imaging centres or hospitals. Particularly beneficial in emergency situations or areas with limited access to medical imaging services, mobile medical imaging ensures more efficient and accessible radiology services.

Apart from its use in remote areas, the technologies can be used in long-term care facilities, nursing homes, and outpatient clinics, paving the way for rapid, convenient, and cost-effective radiology services across diverse settings. As an added benefit, this year, new helium-free mobile imaging technology has entered the market. Phillips introduced the BlueSeal MR Mobile in November, the industry’s first and only 1.5T fully sealed helium-free magnet mobile unit that addresses resource constraints and sustainability issues associated with conventional scanners.

“Having access to mobile MRI scanners is a real game-changer for remote and rural communities around the world,” said Ruud Zwerink, General Manager, Magnetic Resonance at Philips at the launch event. “By introducing our breakthrough technology into the mobile MR market, we are significantly reducing the operational and sustainability issues associated with conventional scanners and helping healthcare providers to deliver fast, patient-friendly quality of care.”

Revolutionising access to imaging data

Web-based enterprise imaging systems are replacing traditional picture archiving and communication systems (PACS), eliminating siloes between modalities. Clinicians can now access images and reports from anywhere without the need for specific workstations. Integration of AI and advanced imaging tools into these systems facilitates seamless interaction with electronic medical records, providing greater access to images and reports across health systems and enabling sharing with patients.

Third-party server farm firms like Google Health and Amazon are driving the shift towards cloud-based archive storage. In 2022, Amazon introduced HealthLake Imaging, a HIPAA-eligible capability aimed at addressing challenges in managing the increasing volume and complexity of medical imaging data to simplify storage, access, and analysis of medical images at a petabyte scale. HealthLake Imaging is estimated to reduce the total cost of medical imaging storage by up to 40 per cent.

Hospitals recognise the benefits of outsourcing storage to companies with around-the-clock monitoring and the flexibility to scale storage without the need for additional hardware. This approach proves to be more economically viable, freeing hospitals from the burden of maintaining large IT storage areas and support infrastructure.

Transforming diagnosis with POCUS

Point-of-care ultrasound (POCUS) has revolutionised medical imaging, providing real-time images that enable physicians to assess organ function, evaluate cancer risk, and diagnose various medical conditions quickly and accurately. The pandemic has underscored the importance of POCUS, particularly in diagnosing and monitoring COVID-19 patients. The development of portable, handheld POCUS devices further minimises the risk of disease transmission.

With applications in emergency medicine, critical care, cardiology, and obstetrics-gynaecology, POCUS is poised to become a standard triage tool, offering detailed and accurate information to guide clinicians in making informed decisions about patient care.

In Abu Dhabi, Sheikh Shakhbout Medical City (SSMC) became the first in the Middle East to launch an academy for upskilling medical practitioners using an AI-guided POCUS device in 2022. “This multidisciplinary course is important as it improves the initial assessment process, advances the timelines and quality of care for patients and, ultimately, saves more lives,” says Dr. Siddiq Anwar, a consultant nephrologist at SSMC.

The rise of hyperspectral and molecular imaging

Hyperspectral and molecular imaging technologies are on the rise, driven by the demand for more detailed and accurate diagnostic information. Hyperspectral imaging captures images at multiple wavelengths, facilitating the identification and analysis of specific tissues or substances within the body. Molecular imaging, utilising targeted probes, visualises specific molecular targets.

Examples like X-ray spectroscopy (XS) and micro-CT showcase the traction gained by hyperspectral and molecular imaging in the medical field. XS, a non-invasive imaging technique, offers high-resolution information about the elemental composition of tissues and organs, enhancing the accuracy of diagnosis. Micro-CT, a high-resolution imaging modality, uses X-rays to produce detailed images of small structures, such as bone microarchitecture and small tumors.

These advanced imaging technologies surpass traditional X-ray imaging, providing higher resolution, greater specificity, and increased sensitivity. Consequently, clinicians gain a more accurate and detailed understanding of the body’s internal structures and functions, enabling earlier detection and more targeted treatment of diseases.

The emergence of photon-counting CT marks a future wave for CT imaging systems, promising reduced radiation dose, improved image quality, and built-in spectral imaging capabilities.

In the case of detecting congenital heart defects among infants, a new study reveals that photon-counting computed tomography (PCCT) offers better cardiovascular imaging quality at a similar radiation dose, compared to dual-source CT (DSCT). More than 97 per cent of the PCCT images were at least diagnostic quality, compared to 77 per cent of the DSCT images.

“Infants and neonates with suspected congenital heart defects are a technically challenging group of patients for any imaging method, including CT,” says Dr.Timm Dirrichs, senior physician and specialist in cardiothoracic radiology at RWTH Aachen University Hospital, Germany. To effectively plan for surgery and generate virtual and printed 3D reconstructions of the heart, a thorough assessment involving ultrasound, MRI, and CT exams is typically required.

“PCCT is a promising method that may improve diagnostic image quality and efficiency compared to DSCT imaging,” Dr. Dirrichs adds. “This higher efficiency can be used to reduce the radiation dose at a given image quality level or to improve image quality at a given radiation level.”

Maximising the potential of MRI

Magnetic resonance imaging (MRI) has become indispensable in modern medicine, offering high-resolution images of the body’s internal structures. Anticipated advancements in 2024 focus on developing more powerful magnets for higher-resolution images in less time. This not only enhances diagnostic accuracy but also improves patient comfort by reducing scan times.

In the pursuit of more cost-effective MRI technology, artificial intelligence (AI) and machine learning algorithms play a pivotal role. These technologies identify artifacts and noise in images, allowing real-time adjustments during scans and leading to more accurate diagnoses.

This year, a multi-institutional Adolescent Brain Cognitive Development Study used deep learning AI to identify markers of ADHD. This study combined brain imaging and clinical surveys, incorporating specialised MRI data, diffusion-weighted imaging (DWI).

“ADHD often manifests early and profoundly impacts one’s quality of life and societal functioning,” says study co-author Justin Huynh, a research specialist in the Department of Neuroradiology at the University of California, San Francisco, and a medical student at the Carle Illinois College of Medicine at Urbana-Champaign.

“There is definitely an unmet need for more objective metrics for diagnosis. That’s the gap we are trying to fill.” The study’s findings, published in November, offer promise for an objective diagnostic method for a condition impacting 129 million children and adolescents globally.

3D imaging goes mainstream

A study involving over a million women revealed that digital breast tomosynthesis (DBT) outperforms standard digital mammography in breast cancer screening outcomes. The cancer detection rate with DBT was higher at 5.3 per 1,000 screened, compared to 4.5 per 1,000 screened with 2D digital mammography alone. Additionally, DBT demonstrated a lower rate of false positives and recalls during screening.

Beyond DBT, mammography is rapidly transitioning to 3D tomosynthesis systems, constituting nearly 50 per cent of breast imaging systems in the US, according to FDA data. Despite the additional read time and increased archive storage space, 3D systems offer advantages such as reduced false positives, fewer unnecessary biopsies, and improved assessments for radiologists by allowing them to examine slices of breast tissue.

As we look to the future, the ongoing growth and development in medical imaging technology promise clinicians more powerful and effective tools for disease diagnosis and treatment. By fully harnessing these technologies, we can anticipate improved patient outcomes, enhanced care quality, and increased accessibility and cost-effectiveness of healthcare services for all.

Back to Technology

Mayo Clinic researchers recently invented a new class of artificial intelligence (AI) algorithms called hypothesis-driven AI that are a significant departure from traditional AI models that learn solely from data.

In a review published in Cancers, the researchers note that this emerging class of AI offers an innovative way to use massive datasets to help discover the complex causes of diseases such as cancer and improve treatment strategies.

"This fosters a new era in designing targeted and informed AI algorithms to solve scientific questions, better understand diseases, and guide individualised medicine," says senior author and co-inventor Hu Li, Ph.D., a Mayo Clinic Systems biology and AI researcher in the Department of Molecular Pharmacology and Experimental Therapeutics. "It has the potential to uncover insights missed by conventional AI."

Conventional AI is primarily used in classification and recognition tasks, such as face recognition and imaging classification in clinical diagnosis, and it has been increasingly applied to generative tasks, such as creating human-like text. Researchers note that conventional learning algorithms often do not incorporate existing scientific knowledge or hypotheses. Instead, these rely heavily on large, unbiased datasets, which can be challenging to obtain.

Dr. Hu Li, Ph.D., a Mayo Clinic Systems biology and AI researcher in the Department of Molecular Pharmacology and Experimental Therapeutics.

According to Dr. Li, this limitation considerably restricts the flexibility of AI methods and their uses in areas that demand knowledge discovery, like medicine.

AI is a valuable tool for identifying patterns in large and complex datasets like those employed in cancer research. The central challenge in using conventional AI has been maximising the embedded information within those datasets.

With hypothesis-driven AI, researchers look to find ways to incorporate an understanding of a disease, for example, integrating known pathogenic genetic variants and interactions between certain genes in cancer into the design of the learning algorithm. This will enable researchers and clinicians to determine which components contribute to model performance and, hence, enhance interpretability. Further, this strategy can address dataset issues and promote our focus on open scientific questions.

"This new class of AI opens a new avenue for better understanding the interactions between cancer and the immune system and holds great promise not only to test medical hypotheses but also predict and explain how patients will respond to immunotherapies," says Daniel Billadeau, Ph.D., a professor in Mayo Clinic's Department of Immunology. Billadeau is a co-author and co-inventor of the study and has a long-standing research interest in cancer immunology.

The research team says hypothesis-driven AI can be used in all sorts of cancer research applications, including tumour classification, patient stratification, cancer gene discovery, drug response prediction and tumour spatial organisation.

Dr. Li notes that the disadvantage of this tool is that creating these types of algorithms requires expertise and specialised knowledge, potentially limiting wide accessibility. There is also potential for building in bias, and they say researchers must watch for that when applying different pieces of information. In addition, researchers generally have a limited scope and won't be formulating all possible scenarios, potentially missing some unforeseen and critical relationships.

"Nonetheless, hypothesis-driven AI facilitates active interactions between human experts and AI, that relieve the worries that AI will eventually eliminate some professional jobs," Dr. Li says.

Since hypothesis-driven AI is still in its infancy, questions remain, such as how to best integrate knowledge and biological information to minimise bias and improve interpretation. Dr. Li says despite the challenges, hypothesis-driven AI is a step forward.

Back to Techology

startAD, the Abu Dhabi-based startup accelerator powered by Tamkeen and anchored at NYU Abu Dhabi (NYUAD), and the Department of Health – Abu Dhabi (DoH), the regulator of the healthcare sector in the Emirate, today signed an agreement to launch HealthX. A startup programme that seeks to advance Abu Dhabi as a hub in the biotech and life science industry, HealthX will advance 30 global cutting-edge life science startups into the UAE’s healthcare ecosystem over two years.

The agreement was signed by Dr. Asma Ibrahim Al Mannaei, Executive Director of Research and Innovation Centre at the Department of Health Abu Dhabi (DoH) and Managing Director of startAD Ramesh Jagannathan at the Abu Dhabi Global Healthcare Week (ADGHW), which was held earlier this week. The timing of the announcement underscores both parties’ commitment to helping boost the UAE’s innovation credentials in the healthcare sector.

Applications are now open for the inaugural cohort on generative AI in healthcare and life sciences. Subsequent cohorts will focus on genomics and precision medicine, neurodegenerative disease, advanced therapeutics, drug discovery, mental health, and assistive technology for People of Determination.

The incubator offers six weeks of intensive training, personalized support, and the potential to secure customers in the UAE. It assists global and local entrants develop a highly customized pilot proposal and a deep understanding of the UAE market. Successful applicants will benefit from access to the DoH regulatory sandbox and UAE de-identified data to help refine their product, along with admittance to labs and opportunities to collaborate on cutting-edge research at leading academic institutes. The business support clinic includes finance, legal, licensing, IP support, ongoing mentorship, and introduction to venture capitalists. Networking prospects include pilot discussions with healthcare partners, licensing opportunities from the DoH, along with an interconnected ecosystem with key players such as DoH, NMC Healthcare, Burjeel Holdings and Harley Street Medical Centre, and onboarding to the startAD alumni programme.

“As the regulator of the healthcare sector in the Emirate, DoH continues to spearhead borderless collaborations, research and innovation to enable a healthier future for all, hence positioning Abu Dhabi as a leading destination for healthcare and life sciences on the global stage. We are delighted to be collaborating with startAD, a partnership poised to tackle global healthcare challenges through supporting start-ups and enabling innovation,” said Dr. Al Mannaei.

Jagannathan further added: “The future of health holds immense possibilities, and Abu Dhabi aims to be a global leader in the field. In partnership with the Department of Health, we are leading the charge to transform the industry and co-create a sustainable operating environment for global healthcare life sciences startups. Our focus is on attracting innovators dedicated to propelling the UAE to the forefront of healthcare, life-science research, and innovation. With top-quality services and a commitment to excel in healthtech and biotech, I believe we are setting the stage for something extraordinary.”

The initiative is open to global and local pre-seed to series B startups developing innovative technologies with a clear use case for the Abu Dhabi healthcare industry. A strong team will include at least two members with relevant experience, demonstrated traction through existing pilots and customers, and a willingness to expand into the UAE, if not already established in the country. Startups with a product or minimum viable product (MVP) ready to test and implement with corporate customers and a clear roadmap showcasing scalability potential are encouraged to apply.

Life sciences and biotechnology, combining the power of biology, genetics, and technology, are rapidly developing fields that show great promise in addressing pressing health challenges today and in the future. The technologies can also be used to develop health products and therapies, from bioengineered tissues to next-generation vaccines, at lower costs than traditional solutions. The Abu Dhabi healthcare market is ripe for advanced technology startups to make a tangible impact with an existing ecosystem of best-in-class healthcare provision, a wealth of data and digital infrastructure, and a globally leading genomics programme. The vision of the programme is to provide a strong platform for startups to address critical challenges in healthcare delivery, improve patient outcomes, and streamline operations across the healthcare value chain.

Back to Management

Cerebral palsy (CP) is one of the most challenging disorders to manage due to its varied manifestations and profound impact on the developmental stages of childhood. For us, as medical professionals, treating CP in children is not just a medical challenge but a commitment to improving lifelong outcomes through early intervention and innovative therapies.

The complexity of CP in children stems from its varying degrees of severity: no two cases are the same, and treatments must be as unique as the individuals we care for. Traditionally, treatment has focused on enhancing function through physical and occupational therapy, as well as managing spasticity with medications. However, our understanding and capabilities are expanding, bringing hope to affected families.

Recent advances in medical research have shown promising developments, particularly in the use of hyperbaric oxygen therapy (HBOT). This treatment, involving the inhalation of pure oxygen in a pressurised environment, has been studied for its potential to improve cognitive and motor abilities in children with CP, and this state-of-the-art hyperbaric oxygen therapy treatment protocol combined with other proven interventions is available at DP World's Aviv Clinics. Our clients have had significant improvements in the symptoms of cerebral palsy in the cases we have treated. Studies suggest that HBOT can enhance brain plasticity, potentially improving motor functions and cognitive processing by fostering new neural pathways.

Moreover, the role of emerging technologies and therapies is increasingly significant. Innovative neurorehabilitation approaches, such as robotic-assisted therapies and virtual reality, are being explored for their potential to provide engaging, effective treatment modalities for young patients. These technologies aim to improve physical outcomes and enrich the therapy experience for children by making it more interactive.

The investment in advanced imaging techniques and genetic testing is also crucial. These tools can help us better understand the etiology of CP in each patient and tailor interventions more precisely, enhancing the efficacy of treatments from an early age. Such personalised medicine approaches are the future of care in CP, promising interventions aligned with individual genetic profiles and specific neurological pathways.

The economic implications of investing in advanced treatments for paediatric CP are profound. Early and effective intervention can reduce the lifetime cost of care per child by mitigating some of the severe complications that can arise. From a business perspective, investing in advanced, efficacious treatments aligns with our goals of cost-effectiveness and enhances patient outcomes, which is paramount.

As we move forward, it is essential to focus on comprehensive care strategies that address the physical symptoms of CP and the psychological and social challenges these young patients face. Incorporating family counselling and support services into the treatment regime is crucial for providing holistic care.

The treatment of cerebral palsy in children is evolving rapidly, driven by scientific advances and a deeper understanding of the condition's broader impacts. As healthcare professionals, we must continue to advocate for and invest in developing treatments that offer meaningful improvements in the lives of these young patients and their families. Our collective efforts in embracing and implementing these advanced therapies will change the trajectory of paediatric CP treatment and underscore our commitment to the next generation's health and well-being.

Dr. Zemer Wang is the Medical Director at DP World’s Aviv Clinics.

Back to Clinical