Artificial Intelligence is revolutionizing many aspects of life, including how doctors approach mesothelioma diagnosis, treatment, and patient care. Advanced computational technologies are improving diagnostic accuracy, speeding up investigational processes, and personalizing treatment approaches. From predicting patient outcomes to discovering novel therapies, AI is transforming mesothelioma care in ways that were unimaginable just a few years ago and providing new hope for patients with this rare, asbestos-related cancer.
What is Artificial Intelligence?
Artificial Intelligence (AI) refers to computer systems designed to perform tasks that typically require human intelligence. Artificial Intelligence can be used for everything from learning from data, recognizing patterns, making decisions, and solving complex problems.[1] In healthcare, AI encompasses several types of technologies, including machine learning, deep learning, and neural networks.
Machine learning is a subset of AI designed to help computers learn and improve from human experience and knowledge, without being explicitly programmed. Deep learning, a more advanced form of machine learning, uses artificial neural networks to process vast amounts of data and identify intricate patterns. When applied to healthcare, these technologies can analyze medical images, patient records, genetic information, and research data at speeds and levels of complexity that go far beyond human ability.[2]
When used for mesothelioma research, care, and treatment, AI systems can process information from thousands of patient cases, medical images, and research studies to identify patterns that help doctors make better decisions about diagnosis, treatment selection, and predicting patient outcomes. The technology continues to evolve, offering increasingly sophisticated tools for managing this challenging disease.
How is Artificial Intelligence Used by Healthcare Professionals and Researchers?
Healthcare professionals and researchers are leveraging AI across multiple aspects of mesothelioma care, from initial research and drug discovery to diagnosis and treatment planning. These applications are fundamentally changing how medical teams approach this rare cancer.
Research
AI is accelerating the pace of mesothelioma research in remarkable ways. Scientists are using computational platforms powered by AI to identify potential drug candidates much faster and more efficiently than traditional laboratory methods ever have.[3]
- Insilico Medicine, a Hong Kong-based biotechnology company, has pioneered the use of generative AI in developing treatments for mesothelioma. Their Pharma.AI platform, which combines genomics, big data analysis, and deep learning for drug discovery, was used to develop a drug that works by blocking the transcription of genes that promote cancer progression, potentially preventing tumor cells from proliferating and surviving. It accelerated the identification of the compounds that mesothelioma cells respond to, significantly reducing the time needed for drug development. The result was ISM6331, a drug that received FDA Investigational New Drug clearance and orphan drug designation, allowing the company to begin Phase I clinical trials. This represents just one example of how AI is shortening the timeline from discovery to patient access for promising new therapies.
- Researchers are also using machine learning to better understand the gut microbiome’s role in mesothelioma treatment response. UK researchers analyzing data from a clinical trial used advanced machine learning techniques to identify specific bacterial species that differ between patients who respond well to immunotherapy and those who don’t. This discovery suggests that bacterial balance in the gut creates what researchers call a “rheostat effect,” potentially controlling how effectively immunotherapy works by shaping the immune environment around tumors.
- AI is also helping scientists create computer models to improve existing treatments. Researchers at the University of Surrey and GSI Helmholtzzentrum für Schwerionenforschung used Artificial Intelligence to create a sophisticated computer model of human lung tissue. This model can be used to accurately simulate how radiation will impact the tissue of specific mesothelioma patients on a cell-by-cell basis, helping doctors determine and tailor the optimal radiation dose for each individual.
Diagnostics
Diagnosing mesothelioma remains challenging due to the disease’s rarity, complexity, and similarity to other conditions, so one of the AI applications that researchers are most excited about is its ability to improve diagnostic accuracy and speed through sophisticated pattern recognition and analysis.[3]
- Researchers from Scotland’s University of Glasgow used Artificial Intelligence to analyze 3,446 whole-slide images from resected mesothelioma tumor samples. The result was the identification of 16 recurrent patterns in tissue samples that correlate with patient outcomes. The group then created a comprehensive histomorphological atlas of the disease that provides doctors with an unprecedented tool that provides accurate diagnoses and helps distinguish mesothelioma from similar conditions. This outcome addressed one of the biggest challenges in mesothelioma diagnosis: the subjective nature of examining stained tissue images and the lack of standardized protocols for interpreting complex and sometimes conflicting tissue characteristics.
- A comprehensive analysis by Iranian researchers reviewed the outcomes of nearly 200 studies on AI applications in mesothelioma management and found that AI demonstrated superior predictive abilities for airway disorders compared to pulmonologists working alone. They also found that, collectively, the applications had remarkable accuracy rates, with support vector machine models achieving 99.97% accuracy in mesothelioma prediction and random forest models reaching 93.75% accuracy. Even more impressively, logistic regression models used in two studies achieved 100% accuracy.
These diagnostic tools offer significant advantages over traditional methods. AI can distinguish between malignant and benign cell proliferation, identify disease subtypes, and reduce the time needed for diagnosis—all while being less expensive than invasive procedures like thoracoscopy and laparoscopy.
Treatment
AI is transforming how doctors select and deliver treatments for mesothelioma patients. By analyzing vast amounts of patient data, AI systems can predict which treatments are most likely to benefit individual patients and help doctors avoid therapies that have proven ineffective in patients with similar characteristics.[3]
- Memorial Sloan Kettering Cancer Center researchers developed OncoCast-MPM, a machine-learning risk prediction model designed specifically for use in mesothelioma patients. The system analyzes clinical, pathological, and molecular data—incorporating 274 variables—to provide individualized risk scores for each patient. It accurately separates patients by survival outcomes as early as one year into diagnosis, without requiring special testing. The accurate, personalized prognoses offered by the platform is essential to providing individualized patient care, prioritizing high-risk patients for clinical trials, and creating matched historical control groups for research.
- Scientists at Cedars-Sinai Cancer Center have taken personalization even further by using AI to create “molecular twins“—virtual replicas of patient tumors built from clinical data combined with multi-omic molecular data. This approach helps identify the best therapies and predicts how cancer will impact each patient. The platform relies on thousands of previously-programmed data points to provide more comprehensive predictions than traditional methods, which have only focused on cancer type or stage.
- A computer lung tissue model developed by University of Surrey researchers exemplifies how AI enables truly targeted treatment. By examining radiation’s interaction with tissue at the cellular level, this model helps physicians choose exactly the right length and strength of radiotherapy tailored to each patient’s specific biology.
Looking ahead, researchers predict that AI will increasingly create precision medicine approaches that match patients with therapies based on their unique genetic and molecular profiles. It has the potential to transform mesothelioma from an untreatable disease into one with reasonable long-term treatment options.
How Can Patients Use Artificial Intelligence?
While most AI applications in mesothelioma care are applied by healthcare professionals, patients are increasingly benefiting from their own use of AI-powered tools. Many have found that Artificial Intelligence helps them ensure they receive the best care informed by the latest AI advances.
Accessing AI-Enhanced Care
Patients seeking mesothelioma treatment should ask their medical teams whether AI-powered diagnostic tools, prognostic models, or treatment planning systems are available at their treatment center. Many leading cancer centers now incorporate AI into their standard workflows, though patients may need to advocate for themselves to ensure they have access to these technologies.[4]
When considering clinical trials, patients should inquire whether the trial incorporates AI-driven drug discovery or selection methods. Treatments developed through AI platforms may offer cutting-edge therapies and treatment options that physicians would not have chosen through traditional approaches.
Understanding Personalized Predictions
Patients can benefit from AI-generated prognostic information that helps them understand their individual situation more clearly. Tools like OncoCast-MPM provide risk assessments based on hundreds of variables specific to each patient, offering more nuanced predictions than traditional staging systems alone.
However, patients should remember that AI predictions are meant to inform decisions based on probabilities and are not absolute predictions of individual outcomes. These tools work best when used in conjunction with the expertise and judgment of experienced mesothelioma specialists.
Lifestyle Considerations
Research into the gut microbiome’s role in immunotherapy response suggests that patients’ dietary choices may influence treatment outcomes. Though scientists are still investigating this connection, patients undergoing immunotherapy may want to ask their physician whether dietary modifications to promote beneficial gut bacteria could potentially improve their response to treatment, especially since high-fiber diets that feed beneficial bacteria have shown promise in other cancer types.
Staying Informed
Patients can stay current on AI developments in mesothelioma care by following reputable cancer research organizations, asking their doctors about new AI-powered tools, and connecting with patient advocacy groups that track emerging technologies. Understanding these advances helps patients engage in informed discussions with their care teams about treatment options.
What are the Drawbacks to Using Artificial Intelligence?
Despite its tremendous promise, AI faces several important limitations and challenges, and these are especially important for mesothelioma patients and their doctors to understand. Concerns include:
Data Quality and Availability
AI systems are only as good as the data they’re trained on, and because mesothelioma is so rare, the datasets available for training AI models are relatively small compared to more common cancers. This limited data can affect the accuracy and ability to make generalizations with AI predictions. This is especially true for the rarer subtypes of mesothelioma, and for patients with unusual characteristics or co-morbidities not well-represented in training data.
Additionally, AI models trained primarily on data from one population may not perform as well when applied to patients with different demographic characteristics, genetic backgrounds, or environmental exposures. This raises concerns about equity and ensuring that AI benefits all patients equally.[]
Lack of Standardization
Different AI systems use different methodologies, making it difficult to compare results across studies or choose between competing approaches. The field lacks standardized protocols for developing, validating, and implementing AI tools in patient populations. This variability can create confusion for doctors trying to decide which AI systems to trust and use.
Limited Understanding of Decision-Making
Many AI systems, and especially those using deep learning, function as “black boxes.” They can produce highly accurate predictions but may not be able to explain the basis for their conclusions. This lack of transparency can make it difficult for doctors to understand why an AI system made a particular recommendation or to assess whether the information the system provides might be a mistake.
Cost and Access Issues
Implementing AI systems requires significant investment in technology infrastructure, training, and ongoing maintenance. Smaller or rural facilities aren’t likely to have the resources to adopt these technologies, creating disparities in care quality between patients who have access to AI-enhanced treatment and those who don’t.[5]
Regulatory and Validation Challenges
AI tools must undergo rigorous validation to ensure they’re safe and effective before being used in clinical practice. The regulatory landscape for AI in healthcare is still evolving, and the process for approving new AI applications takes a long time. Additionally, as new data becomes available, AI systems will need to be updated, raising questions about how frequently these tools should be revalidated.
Not a Replacement for Clinical Expertise
AI should be viewed as a tool to augment, not replace, the judgment and expertise of experienced mesothelioma specialists. As good as it is, the technology cannot account for all the nuances of an individual patient’s situation, values, and preferences. The human element of medicine—including the doctor-patient relationship and individualized decision-making—is irreplaceable.
Potential for Over-Reliance
There’s a risk that doctors could place too much faith in AI predictions without adequately considering their limitations or applying critical thinking to the results offered. Conversely, some patients might become anxious about AI-generated risk predictions without understanding that these represent probabilities across populations, not certainties for individual patients.
Privacy and Data Security
AI systems need access to detailed patient information, including genetic data and medical histories. Protecting this sensitive information from breaches while still enabling the data sharing necessary for AI development presents an ongoing challenge.[6]
AI Offers Hope for Better Outcomes
Despite these limitations, the field continues to advance rapidly. Ongoing research focuses on addressing these challenges through improved algorithms, larger and more diverse datasets, better validation methods, and frameworks for responsible AI implementation. As these issues are resolved, AI’s role in mesothelioma care is likely to expand significantly.
For patients and families navigating mesothelioma, understanding both the promise and limitations of AI helps set realistic expectations while remaining open to the genuine improvements these technologies are bringing to care. The combination of cutting-edge AI tools with human expertise offers the best hope for advancing treatment and improving outcomes for this challenging disease.
References
View All References
