Scribd
Upload
5 views4 pages

AI Bayesian Algorithm for Bile Duct Cancer

This document presents a Bayesian probability-based algorithm for diagnosing, planning treatment, and prognosing bile duct cancer, integrating various risk factors and biomarkers. The algorithm includes mathematical formulations, worked examples, and treatment recommendations based on expected utility. The authors argue that this approach can enhance diagnostic confidence and optimize treatment decisions, ultimately improving patient outcomes.

Uploaded by

Dale Srinivas
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
5 views4 pages

AI Bayesian Algorithm for Bile Duct Cancer

This document presents a Bayesian probability-based algorithm for diagnosing, planning treatment, and prognosing bile duct cancer, integrating various risk factors and biomarkers. The algorithm includes mathematical formulations, worked examples, and treatment recommendations based on expected utility. The authors argue that this approach can enhance diagnostic confidence and optimize treatment decisions, ultimately improving patient outcomes.

Uploaded by

Dale Srinivas
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Artificial Intelligence: A Bayesian

Probability Principle-Based Algorithm


for Bile Duct Cancer and Treatment
Authors:
Dale Srinivas*, Sarika Gupta (NHS, UK), Anu Singh, Arvind Kumar Gupta (NHS, UK)

Affiliations:
Twin Island University, Bushlot, Guyana (Dale Srinivas, Anu Singh)
NHS, UK (Sarika Gupta, Arvind Kumar Gupta)

Abstract
Bile duct cancer (cholangiocarcinoma) is an uncommon but aggressive malignancy.
Artificial intelligence with Bayesian probability offers a structured way to integrate risk
factors, biomarkers, imaging, and pathology. We present a Bayesian network–based
algorithm for diagnosis, treatment planning, and prognosis in bile duct cancer, including
mathematical formulations, worked examples, tables, and illustrative figures.

1. Mathematical Framework
Bayesian Posterior Probability:
Posterior Odds = Prior Odds × ∏ LR(Ei)
P(Ca | E) = Posterior Odds / (1 + Posterior Odds)

Inputs:
- Prior: PSC, liver flukes, chronic biliary inflammation
- Evidence: CA19-9, MRCP, biopsy
- Treatment: resection, transplant, chemotherapy, targeted drugs

1.1 Worked Example


Prior P0=0.20 ⇒ Odds0=0.25
CA19-9 positive LR+=5 ⇒ Odds1=1.25 ⇒ P1=0.56
MRCP stricture LR+=7 ⇒ Odds2=8.75 ⇒ P2=0.86
Biopsy positive LR+=15 ⇒ Odds3=131.25 ⇒ P3=0.99
2. Results
Figure 1. Bayesian Updating Curve

Figure 2. Progression-Free Survival Over Time

Figure 3. Treatment Algorithm Flowchart


Table 1. Likelihood Ratios for Diagnostic Evidence

Test/Marker LR+ (Positive) LR- (Negative)

CA19-9 >100 U/mL 5.0 0.5

MRCP stricture 7.0 0.6

Endoscopic biopsy positive 15.0 0.2

Table 2. Expected Utility-Based Treatment Recommendations

Posterior P(Resectable) Treatment Expected Utility

<20% Palliative chemotherapy 0.3

20–50% Chemotherapy ± stenting 0.5

50–80% Surgery ± adjuvant 0.7

>80% Surgery or transplant 0.9

3. Discussion
Bayesian AI integrates biomarkers, imaging, and pathology into a unified diagnostic and
therapeutic pathway. Sequential evidence updating increases diagnostic confidence, while
expected utility calculations personalize treatment. This approach may reduce delays,
optimize surgical selection, and improve survival outcomes.
4. Conclusion
Bayesian frameworks offer a transparent, rigorous way to guide clinical management of bile
duct cancer. Integration into AI platforms can support clinicians in complex decision-
making, leading to more precise and individualized care.

References
Banales, J.M., Cardinale, V., Carpino, G. et al. (2016) ‘Cholangiocarcinoma: current
knowledge and future perspectives consensus statement from the European Network for
the Study of Cholangiocarcinoma’, Nat Rev Gastroenterol Hepatol, 13, pp. 261–280.

Bridgewater, J., Galle, P.R., Khan, S.A. et al. (2014) ‘Guidelines for the diagnosis and
management of intrahepatic cholangiocarcinoma’, J Hepatol, 60(6), pp. 1268–1289.

Pearl, J. (1988) Probabilistic Reasoning in Intelligent Systems: Networks of Plausible


Inference. San Francisco: Morgan Kaufmann.

Rizvi, S. and Gores, G.J. (2013) ‘Pathogenesis, diagnosis, and management of


cholangiocarcinoma’, Gastroenterology, 145(6), pp. 1215–1229.

You might also like