Diagnosis of Pulmonary Embolism Using Fuzzy Inference System

1
Diagnosis of Pulmonary Embolism Using Fuzzy
Inference System

• Research Assistant Vishwanath Acharya
• Research Director Dr. Gursel Serpen
• Medical Expertise Drs. Parsai, Coombs
  Woldenberg of Medical College of Ohio

2
Why Artificial Intelligence???

• It can offer a competent second opinion.
• It offers the expertise of an expert radiologist
  in interpreting scans when an expert radiologist
  is not available.
• It has the ability to make accurate and quick
  diagnosis.
• It has the potential to reduce inter-observer
  variability.

3
Artificial Intelligence in Practice
4
Groups Facing Higher Probability of Pulmonary
Embolism

• Patients Undergoing various types of surgery -
  general, urological, neuro-surgical, and
  gynecological.
• Patients with orthopedic problems and chronic
  diseases.

• These groups face a higher probability of
  Pulmonary Embolism due to the high risk of
  developing deep venous thrombosis.

5
Various Diagnostic Criterias

• PIOPED - Prospective Investigation of Pulmonary
  Embolism Diagnosis 1995.
• Biellos Criteria 1979.
• Inputs from Expert Radiologists.

6
PIOPED Criteria

• Low Probability
• Multiple Matching V/Q defects.
• Corresponding V/Q defects and CXR parenchymal
  opacity in upper or middle lung zone.
• Corresponding V/Q defects and large Pleural
  Effusion.
• gt 3 Small SPD.
• Very Low Probability
• lt 3 Small SPD.
• Normal
• No perfusion defects and perfusion outlines the
  shape of the lung seen on CXR

• High Probability
• gt 2 Large segmental perfusion defects (SPD).
• 1 Large SPD and gt 2 Moderate SPD.
• gt 4 Moderate SPD.
• Intermediate Probability
• 1 Moderate to lt 2 Large SPD.
• Corresponding V/Q defect and CXR opacity in lower
  lung.
• Single moderately matched V/Q defect.
• Corresponding V/Q defect and small Pleural
  Effusion.

7
Biello-Siegel Criteria

• High Probability (87 )
• 2 Large gt 75 of a segment mismatches.
• Q defect gtgt CXR and V.
• Intermediate Probability (20 - 33 )
• Abnormality not within low or high category.
• Low Probability (10 )
• Small lt 25 of a segment Q defect(S).
• Matched V/Q defects involving lt 1/3 of lung.
• Non-Segmental defects.
• Q defect ltlt CXR defect.
• Normal (0 ).

8
Is Fuzzy Logic really Fuzzy? Why Fuzzy Logic?

• Despite its name Fuzzy Logic is not nebulous,
  cloudy or vague.
• It provides a very precise approach for dealing
  with uncertainty which is derived from complex
  human behavior.
• Fuzzy Logic is so powerful, mainly because it
  does not require a deep understanding of a system
  or exact and precise numerical values.
• It uses abstraction that in human beings is
  arrived at from experience or intuition.
• It allows intermediate values and representation
  of knowledge with subjective concepts to be
  defined between conventional evaluation.
• It basically pays attention to the excluded
  middle gray areas.
• It attempts to apply a more human like way of
  thinking in programming of computers.

9
Fuzzy Inference System

• The three major components of the Fuzzy Inference
  System are
• Fuzzifier - Converts the crisp input into
  appropriate fuzzy quantity.
• Inference Engine - Allows the application of the
  rule base to the input parameters whereby
  producing the output.
• Defuzzifier - Converts the output produced by the
  Inference Engine into user understandable terms.

10
Inputs to Fuzzy System(According to PIOPED
Criteria)

• Number of Segmental Perfusions.
• Number of Non-Segmental Perfusions.
• Ventilation/Perfusion Mismatch.
• Chest X-Ray Abnormality.
• Presence of Pleural Effusion.

11
Inputs to Fuzzy System(According to PIOPED
Criteria)

• Wt - Weight (pre-calculation of segmental and
  non-segmental perfusion defects.
• Vqdef - Ventilation-Perfusion Defect Mismatch.
• Cxrab - Chest X-ray abnormality.
• Peff - Presence of Pleural Effusion.

12
Rule Base of Fuzzy System(Modeling of the PIOPED
Criteria)
13
Outputs from Fuzzy Inference System(According to
PIOPED Criteria)

• Output of the Fuzzy System models the
  diagnostic capabilities of the Fuzzy System.
  Hence, the various classes are
• Normal.
• Very Low.
• Low.
• Intermediate.
• High
• The output of the Fuzzy System are mapped to
  one of these classes.

14
Outputs from Fuzzy Inference System(According to
PIOPED Criteria)

• Dia - Diagnosis, is the output of the Fuzzy
  System and is divided into 5 classes.
• What you see here is the tweaking that has to
  be given to all the classes in order to implement
  the PIOPED criteria to its best fit.

15
Testing/SimulationTo ensure accuracy and
usability, the software has to pass stringent
tests. These tests were applied in two phases.

• Alpha Testing
• Output data was obtained and passed to
  radiologists to check for accuracy.
• Data developed by radiologists was run through
  the system and checked to ensure that is produced
  expected results.

• Beta Testing
• Currently being implemented. In this phase the
  radiologist will have a hands on experience.
  This will ensure that the software has a high
  degree of usability and physicians wont be
  intimidated by it.

16
Conclusions

• Implementation of Artificial Intelligence
  software in the diagnosis of medical diseases is
  feasible and can be very easily extended to cover
  different diseases.
• It can be of help to medical practitioners.
• The alternative methods utilized to diagnose for
  Pulmonary Embolism effectively capture the spirit
  of the PIOPED criteria.
• This software has the ability to make accurate
  and quick diagnosis.