Artificial Intelligence (AI) and Machine Learning (ML) in early diagnosis of PDAC

Artificial intelligence (AI), a revolutionary technology, is playing a major role in day to day life. We are seeing a lot of evolution in various machine learning (ML) methodologies. AI & ML have become more accurate and applicable to a variety of tasks and are being widely used to solve a whole range of hitherto intractable problems. They have increasingly helping to uncover hidden insights into clinical decision-making, connect patients with resources for self-management, and extract meaning from previously inaccessible, unstructured data assets.

Pancreatic ductal adenocarcinoma (PDAC), a highly malignant tumor and one of the most lethal cancers, characterized by rapid progression, metastasis, and difficulty in diagnosis. Due to the anatomic location of the pancreas, symptoms (weight loss, fatigue, abdominal and back pain, and malaise) and non-availability of an effective testing method at present, the early stage detection has become difficult.

A research team (Guangxi Wang. et. al., 2021) at Institute of Systems Biomedicine, Peking University Health Science Center, Beijing introduce an approach that uses ML and lipidomics to detect PDAC. Metabolomics allows the collection, detection, and analysis of all kinds of small-molecule metabolites, which are highly sensitive to biological activities and pathological conditions. Because of the great metabolite coverage of untargeted metabolomics and reliability of targeted metabolomics, the integration of both assays is a powerful strategy for disease-related biomarker studies. Thus, accurate, robust, and low-cost metabolomics detection methods hold promise for future disease diagnoses. Through greedy algorithm and mass spectrum feature selection, the research team optimized 17 characteristic metabolites as detection features and developed a liquid chromatography-mass spectrometry-based targeted assay. In this study, they sought to combine ML and metabolomics performed with lipid metabolites of serum from patients with PDAC and normal individuals to classify and select lipid features of PDAC. Then, a targeted lipid multiple reaction monitoring (MRM)–mode quantification assay for PDAC detection was established and validated in large sizes of samples.

Using this method, the team studied 1033 patients with PDAC at various stages and achieved 86.74% accuracy with an area under curve (AUC) of 0.9351 in the large external validation cohort and 85.00% accuracy with 0.9389 AUC in the prospective clinical cohort. Accordingly, single-cell sequencing, proteomics, and mass spectrometry imaging were applied and revealed notable alterations of selected lipids in PDAC tissues. The research team is opined that this method would yield an effective, reliable, and accurate minimally invasive approach to PDAC detection.

–  Dr. Anand, R.
Senior Scientific Officer, KSTA

Reference

Guangxi Wang, et. al.,2021. Metabolic detection and systems analyses of pancreatic ductal adenocarcinoma through machine learning, lipidomics, and multi-omics. Science Advances, Vol 7, Issue 52. https://doi.org/10.1126/sciadv.abh2724

MacCurtain, B.M.; Quirke, N.P.; Thorpe, S.D.; Gallagher, T.K. Pancreatic Ductal Adenocarcinoma: Relating Biomechanics and Prognosis. J. Clin. Med. 2021, Vol. 10, Issue 12. https://doi.org/10.3390/jcm10122711

Thakur, G.; Kumar, R.; Kim, S.-B.; Lee, S.-Y.; Lee, S.-L.; Rho, G.-J. Therapeutic Status and Available Strategies in Pancreatic Ductal Adenocarcinoma. Biomedicines, Vol 9, Issue 2. https://doi.org/10.3390/biomedicines9020178