Ali Musani MD, FCCP
University of Colorado School of Medicine
Artificial Intelligence and Interventional Pulmonology
Are we “stepping in GOD’s shoes”?
WABIP Newsletter
J A N U A R Y 2 0 2 2 V O L U M E 1 0 , I S S U E 1
EXECUTIVE BOARD
Hideo Saka, MD
Japan, Chair
Stefano Gasparini,
MD
Italy, Vice-Chair
Silvia Quadrelli, MD
Argenna, Immediate
Past-Chair
David Fielding MD
Australia, Treasurer
Naofumi Shinagawa,
MD
Japan,
Secretary General
Philippe Astoul, MD
France, President
WCBIP 2022
Menaldi Rasmin, MD
Indonesia, President
WCBIP 2024
STAFF
Michael Mendoza
General Manager
Judy McConnell
Administrator
Kazuhiro Yasufuku
Newsleer Editor-in-
chief
P A G E 2
Arcial Intelligence (AI) is not a novel concept. The
word “Arcial Intelligence” was coined by John
McCarthy in 1956, and the rst general-purpose
mobile robot was developed in 1969. The rapid
growth in technology soon created “Super Blue,” the
“supercomputer,” which defeated the world chess
champions.
AI is a type of computer science used to create intel-
ligent machines that recognize human speech and
objects and learn, plan and solve problems like hu-
mans.
AI can be understood in many ways depending on
one’s perspecve, as described below by some
world experts and leaders in the eld.
• AI is just math” - mulple computaons that are
the basis of AI are also where the technology faces
limitaons. Jana Eggers, the CEO of Nara Logics
• AI is just soware. “There’s no bright line sepa-
rang AI soware from any other kind of computer
soware,” Michael Liman, a computer science pro-
fessor at Brown University.
• The potenal of AI lies in its ability to learn, and
its learning from humans. Mikhail Naumov, co-
founder, president, and CSO of Digital Genius.
• Simply put, humans should be focused on teach-
ing machines, so that machines can focus on exe-
cung against jobs that are too big for humans to
process.” J.J. Kardwell, CEO/co-founder of Ever
String
In the eyes of some world experts of technology,
such as Elon Musk, CEO of Tesla, AI is a double-
edged sword. The potenal of AI is fundamentally changing
just about any aspect of our lives is so profound and self-
perpetuang that the threat of AI geng out of human
control makes him say that “AI is a fundamental risk to the
existence of human civilizaon.”
AI in medicine has been growing by leaps and bounds in all
facets, including diagnoscs, therapeucs, research, device
development, and drug development. Watson, the infa-
mous supercomputer, can diagnose thousands of diseases
with extreme expediency and accuracy. Numerous medical
organizaons now use it. Google’s “AI Rena Doctor” can
examine rena scans and diagnose diabec renopathy.
AI has also played a crucial role in the growth of Interven-
onal Pulmonology (IP) over the years. The below examples
will highlight some revoluonary AI-based developments in
IP.
• LungVision system (Body Vision Medical LTD, Israel) is a
novel technology that integrates pre-procedural CT imaging
into augmented uoroscopic images, presenng real-me
visualizaon of the airways and locaon of the pulmonary
lesion during transbronchial navigaon and biopsy. It ena-
bles lesion tracking during breathing movement and im-
proves lesion localizaon and diagnosc yield. LungVision
may provide equivalent diagnosc outcomes to tradional
ENB plaorms at a fracon of the cost.
• Opcal-based navigaon systems (such as SIRIO,
MASMEC S.p.A., Modugno, BA, Italy) perform Lung Thermal
Ablaon (LTA). Procedural planning, monitoring, and lesion
targeng are generally performed with the help of CT.
More recently, the implementaon of C-arm cone-beam CT
(CBCT) technology has introduced a new image guidance
strategy. Navigaon systems emerge as a valid tool to re-
duce procedural mes and administraon of radiaon dos-
es, allowing electromagnec, opcal, or hybrid tracking of
the devices used during intervenons and their real-me
visualizaon in a model obtained from a previously ac-
quired CT scan. In a recent study published in (1) opcal-
based navigaon system, SIRIO was shown to be an e-
cient tool to perform CT-guided LTA, displaying a signicant
reducon (p < 0.001) in the number of required CT scans,
procedure me, and paents’ radiaon exposure.
• A computer-aided diagnosis (CAD) system is a machine
learning texture model for classifying lung cancer subtypes
using preliminary bronchoscopic ndings is a CAD system.
This CAD system can disnguish cancer types to achieve an