4. ESSENTIAL NEED FOR AN EFFICIENT
COMPUTER DIAGNOSTIC PROGRAM
Diagnostic errors occur very frequently; some are disabling or fatal,
resulting in liability for the physician. Unawareness, underreported
5. ESSENTIAL NEED FOR AN EFFICIENT
COMPUTER DIAGNOSTIC PROGRAM
Diagnostic errors occur very frequently; some are disabling or fatal,
resulting in liability for the physician. Unawareness, underreported
Physicians order an inordinate number of futile tests, not rationally
selected according to a probabilistically determined benefit
6. ESSENTIAL NEED FOR AN EFFICIENT
COMPUTER DIAGNOSTIC PROGRAM
Diagnostic errors occur very frequently; some are disabling or fatal,
resulting in liability for the physician. Unawareness, underreported
Physicians order an inordinate number of futile tests, not rationally
selected according to a probabilistically determined benefit
These errors are due to the impossibility to remember or even learn
thousands of diseases with their corresponding names, symptoms,
tests, procedures, causes, mechanisms, etc.
7. ESSENTIAL NEED FOR AN EFFICIENT
COMPUTER DIAGNOSTIC PROGRAM
Diagnostic errors occur very frequently; some are disabling or fatal,
resulting in liability for the physician. Unawareness, underreported
Physicians order an inordinate number of futile tests, not rationally
selected according to a probabilistically determined benefit
These errors are due to the impossibility to remember or even learn
thousands of diseases with their corresponding names, symptoms,
tests, procedures, causes, mechanisms, etc.
Diseases neither learned nor recalled by the physician remain
undiagnosed or misdiagnosed
8. ESSENTIAL NEED FOR AN EFFICIENT
COMPUTER DIAGNOSTIC PROGRAM
Diagnostic errors occur very frequently; some are disabling or fatal,
resulting in liability for the physician. Unawareness, underreported
Physicians order an inordinate number of futile tests, not rationally
selected according to a probabilistically determined benefit
These errors are due to the impossibility to remember or even learn
thousands of diseases with their corresponding names, symptoms,
tests, procedures, causes, mechanisms, etc.
Diseases neither learned nor recalled by the physician remain
undiagnosed or misdiagnosed
Problem compounded by time constraints imposed by health
insurances (managed care medicine)
9. ESSENTIAL NEED FOR AN EFFICIENT
COMPUTER DIAGNOSTIC PROGRAM
Diagnostic errors occur very frequently; some are disabling or fatal,
resulting in liability for the physician. Unawareness, underreported
Physicians order an inordinate number of futile tests, not rationally
selected according to a probabilistically determined benefit
These errors are due to the impossibility to remember or even learn
thousands of diseases with their corresponding names, symptoms,
tests, procedures, causes, mechanisms, etc.
Diseases neither learned nor recalled by the physician remain
undiagnosed or misdiagnosed
Problem compounded by time constraints imposed by health
insurances (managed care medicine)
Computers can store the contents of many medical texts and swiftly
retrieve the needed information anywhere and at any time
10. EXISTING SO CALLED
“COMPUTER DIAGNOSTIC PROGRAMS”
INTERNIST (former Caduceus)
MEDITEL
QMR (Quick Medical Reference)
ILIAD
PKC (Problem-Knowledge Couplers)
ISABEL
DXplain
11. EXISTING SO CALLED
“COMPUTER DIAGNOSTIC PROGRAMS”
INTERNIST (former Caduceus)
MEDITEL
QMR (Quick Medical Reference)
ILIAD
PKC (Problem-Knowledge Couplers)
ISABEL
DXplain
Excellent remainder of unusual diagnoses, but
Unable to pinpoint the disease(s) that actually afflict a specific patient
Considered educational or training tools rather than diagnostic tools
Poorly accepted and seldom used by practicing health care providers
13. Advantages of our novel computer program
It confirms one or more diseases that indeed afflict a specific patient, instead of
displaying a long list of potential diagnoses that compete with each other
14. Advantages of our novel computer program
It confirms one or more diseases that indeed afflict a specific patient, instead of
displaying a long list of potential diagnoses that compete with each other
It rules in diagnoses evoked by clinical data present, but also rules out those diagnoses
that are not viable, based on sensitivity of clinical data absent
15. Advantages of our novel computer program
It confirms one or more diseases that indeed afflict a specific patient, instead of
displaying a long list of potential diagnoses that compete with each other
It rules in diagnoses evoked by clinical data present, but also rules out those diagnoses
that are not viable, based on sensitivity of clinical data absent
It diagnoses concurrent diseases afflicting simultaneously a specific patient
16. Advantages of our novel computer program
It confirms one or more diseases that indeed afflict a specific patient, instead of
displaying a long list of potential diagnoses that compete with each other
It rules in diagnoses evoked by clinical data present, but also rules out those diagnoses
that are not viable, based on sensitivity of clinical data absent
It diagnoses concurrent diseases afflicting simultaneously a specific patient
It calculates accurately the probability of each diagnosis with our novel mini-max
procedure, instead of applying inexact Bayes formula or other methods
17. Advantages of our novel computer program
It confirms one or more diseases that indeed afflict a specific patient, instead of
displaying a long list of potential diagnoses that compete with each other
It rules in diagnoses evoked by clinical data present, but also rules out those diagnoses
that are not viable, based on sensitivity of clinical data absent
It diagnoses concurrent diseases afflicting simultaneously a specific patient
It calculates accurately the probability of each diagnosis with our novel mini-max
procedure, instead of applying inexact Bayes formula or other methods
It gives primary importance to the cost of obtaining each clinical datum. Our algorithm
interprets cost not only as dollar expense, but also considers discomfort and risk of the
procedure. Four overall cost categories are established: No Cost, Small Cost,
Intermediate Cost, and Great Cost
18. Advantages of our novel computer program
It confirms one or more diseases that indeed afflict a specific patient, instead of
displaying a long list of potential diagnoses that compete with each other
It rules in diagnoses evoked by clinical data present, but also rules out those diagnoses
that are not viable, based on sensitivity of clinical data absent
It diagnoses concurrent diseases afflicting simultaneously a specific patient
It calculates accurately the probability of each diagnosis with our novel mini-max
procedure, instead of applying inexact Bayes formula or other methods
It gives primary importance to the cost of obtaining each clinical datum. Our algorithm
interprets cost not only as dollar expense, but also considers discomfort and risk of the
procedure. Four overall cost categories are established: No Cost, Small Cost,
Intermediate Cost, and Great Cost
It anticipates the benefit that each clinical datum would confer to each diagnosis in the
differential diagnosis list, by calculating with our mini-max procedure how much it would
increase or decrease the current probability of the corresponding diagnosis if the data
were found respectively present or absent in the patient
19. Advantages of our novel computer program
It confirms one or more diseases that indeed afflict a specific patient, instead of
displaying a long list of potential diagnoses that compete with each other
It rules in diagnoses evoked by clinical data present, but also rules out those diagnoses
that are not viable, based on sensitivity of clinical data absent
It diagnoses concurrent diseases afflicting simultaneously a specific patient
It calculates accurately the probability of each diagnosis with our novel mini-max
procedure, instead of applying inexact Bayes formula or other methods
It gives primary importance to the cost of obtaining each clinical datum. Our algorithm
interprets cost not only as dollar expense, but also considers discomfort and risk of the
procedure. Four overall cost categories are established: No Cost, Small Cost,
Intermediate Cost, and Great Cost
It anticipates the benefit that each clinical datum would confer to each diagnosis in the
differential diagnosis list, by calculating with our mini-max procedure how much it would
increase or decrease the current probability of the corresponding diagnosis if the data
were found respectively present or absent in the patient.
It selects and recommends the probabilistically calculated best cost-benefit clinical data
to investigate next in the patient at each diagnostic step, to achieve more efficiently and
economically a final diagnosis
21. Advantages of our novel computer program
(continues)
It recommends a set of best cost-benefit clinical data to be investigated simultaneously
in the patient, based on diverse heuristic strategies
22. Advantages of our novel computer program
(continues)
It recommends a set of best cost-benefit clinical data to be investigated simultaneously
in the patient, based on diverse heuristic strategies
It can display selected partial lists that render the diagnostic task more economic and
manageable, without compromising the accuracy of the result. It accomplishes this
through parameters that can be set at diverse values and through abridged output files,
giving the user the choice to select the ones that fit best his or her preferences
23. Advantages of our novel computer program
(continues)
It recommends a set of best cost-benefit clinical data to be investigated simultaneously
in the patient, based on diverse heuristic strategies
It can display selected partial lists that render the diagnostic task more economic and
manageable, without compromising the accuracy of the result. It accomplishes this
through parameters that can be set at diverse values and through abridged output files,
giving the user the choice to select the ones that fit best his or her preferences
It includes complex clinical presentation models in the database, which list associations
among diseases, such as causes, complications, and other relations, precluding
overlooking associated diseases
24. Advantages of our novel computer program
(continues)
It recommends a set of best cost-benefit clinical data to be investigated simultaneously
in the patient, based on diverse heuristic strategies
It can display selected partial lists that render the diagnostic task more economic and
manageable, without compromising the accuracy of the result. It accomplishes this
through parameters that can be set at diverse values and through abridged output files,
giving the user the choice to select the ones that fit best his or her preferences
It includes complex clinical presentation models in the database, which list associations
among diseases, such as causes, complications, and other relations, precluding
overlooking associated diseases
It deals with interactions between concurrent diseases or drugs that mask important
clinical data of the primary disease, precluding the improper ruling out of the
corresponding diagnosis
25. Advantages of our novel computer program
(continues)
It recommends a set of best cost-benefit clinical data to be investigated simultaneously
in the patient, based on diverse heuristic strategies
It can display selected partial lists that render the diagnostic task more economic and
manageable, without compromising the accuracy of the result. It accomplishes this
through parameters that can be set at diverse values and through abridged output files,
giving the user the choice to select the ones that fit best his or her preferences
It includes complex clinical presentation models in the database, which list associations
among diseases, such as causes, complications, and other relations, precluding
overlooking associated diseases
It deals with interactions between concurrent diseases or drugs that mask important
clinical data of the primary disease, precluding the improper ruling out of the
corresponding diagnosis
It proposes an easy way to handle synonyms of diagnoses and clinical data
26. Advantages of our novel computer program
(continues)
It recommends a set of best cost-benefit clinical data to be investigated simultaneously
in the patient, based on diverse heuristic strategies
It can display selected partial lists that render the diagnostic task more economic and
manageable, without compromising the accuracy of the result. It accomplishes this
through parameters that can be set at diverse values and through abridged output files,
giving the user the choice to select the ones that fit best his or her preferences
It includes complex clinical presentation models in the database, which list associations
among diseases, such as causes, complications, and other relations, precluding
overlooking associated diseases
It deals with interactions between concurrent diseases or drugs that mask important
clinical data of the primary disease, precluding the improper ruling out of the
corresponding diagnosis
It proposes an easy way to handle synonyms of diagnoses and clinical data
It is ease to update
27. Advantages of our novel computer program
(continues)
It recommends a set of best cost-benefit clinical data to be investigated simultaneously
in the patient, based on diverse heuristic strategies
It can display selected partial lists that render the diagnostic task more economic and
manageable, without compromising the accuracy of the result. It accomplishes this
through parameters that can be set at diverse values and through abridged output files,
giving the user the choice to select the ones that fit best his or her preferences
It includes complex clinical presentation models in the database, which list associations
among diseases, such as causes, complications, and other relations, precluding
overlooking associated diseases
It deals with interactions between concurrent diseases or drugs that mask important
clinical data of the primary disease, precluding the improper ruling out of the
corresponding diagnosis
It proposes an easy way to handle synonyms of diagnoses and clinical data
It is ease to update
Such an algorithm, if successful in medicine, may represent a more general model of
reasoning; a paradigm of mental structure and functioning applicable to other inexact
disciplines such as law, sociology, politics, defense, or corporate strategy
29. SOCIOECONOMIC ADVANTAGES
OF OUR PROGRAM
1. Reduces significantly the current numerous diagnostic errors and their dismal
consequences
30. SOCIOECONOMIC ADVANTAGES
OF OUR PROGRAM
1. Reduces significantly the current numerous diagnostic errors and their dismal
consequences
2. Saves specialist consultations; important in emergencies and in underdeveloped
areas were specialists are not available
31. SOCIOECONOMIC ADVANTAGES
OF OUR PROGRAM
1. Reduces significantly the current numerous diagnostic errors and their dismal
consequences
2. Saves specialist consultations; important in emergencies and in underdeveloped
areas were specialists are not available
3. Reduces significantly the number of futile tests and procedures, by recommending
the best cost-benefit clinical data to investigate next in a patient. This saves
expense, risk, and discomfort for the patient and saves social medical resources
32. SOCIOECONOMIC ADVANTAGES
OF OUR PROGRAM
1. Reduces significantly the current numerous diagnostic errors and their dismal
consequences
2. Saves specialist consultations; important in emergencies and in underdeveloped
areas were specialists are not available
3. Reduces significantly the number of futile tests and procedures, by recommending
the best cost-benefit clinical data to investigate next in a patient. This saves
expense, risk, and discomfort for the patient and saves social medical resources
4. On the other hand, if best cost-benefit function proves probabilistically that a
diagnostic procedure is absolutely necessary, it will force managed care
organizations and insurers to authorize such procedures, otherwise curtailed for
being expensive
33. SOCIOECONOMIC ADVANTAGES
OF OUR PROGRAM
1. Reduces significantly the current numerous diagnostic errors and their dismal
consequences
2. Saves specialist consultations; important in emergencies and in underdeveloped
areas were specialists are not available
3. Reduces significantly the number of futile tests and procedures, by recommending
the best cost-benefit clinical data to investigate next in a patient. This saves
expense, risk, and discomfort for the patient and saves social medical resources
4. On the other hand, if best cost-benefit function proves probabilistically that a
diagnostic procedure is absolutely necessary, it will force managed care
organizations and insurers to authorize such procedures, otherwise curtailed for
being expensive
5. Facilitates and improves utilization reviews by hospitals, laboratories, and other
providers.
34. SOCIOECONOMIC ADVANTAGES
OF OUR PROGRAM
1. Reduces significantly the current numerous diagnostic errors and their dismal
consequences
2. Saves specialist consultations; important in emergencies and in underdeveloped
areas were specialists are not available
3. Reduces significantly the number of futile tests and procedures, by recommending
the best cost-benefit clinical data to investigate next in a patient. This saves
expense, risk, and discomfort for the patient and saves social medical resources
4. On the other hand, if best cost-benefit function proves probabilistically that a
diagnostic procedure is absolutely necessary, it will force managed care
organizations and insurers to authorize such procedures, otherwise curtailed for
being expensive
5. Facilitates and improves utilization reviews by hospitals, laboratories, and other
providers.
6. Protect health care providers against unfounded malpractice liability
35. SOCIOECONOMIC ADVANTAGES
OF OUR PROGRAM
1. Reduces significantly the current numerous diagnostic errors and their dismal
consequences
2. Saves specialist consultations; important in emergencies and in underdeveloped
areas were specialists are not available
3. Reduces significantly the number of futile tests and procedures, by recommending
the best cost-benefit clinical data to investigate next in a patient. This saves
expense, risk, and discomfort for the patient and saves social medical resources
4. On the other hand, if best cost-benefit function proves probabilistically that a
diagnostic procedure is absolutely necessary, it will force managed care
organizations and insurers to authorize such procedures, otherwise curtailed for
being expensive
5. Facilitates and improves utilization reviews by hospitals, laboratories, and other
providers.
6. Protect health care providers against unfounded malpractice liability
7. Helps to overcome the current well known resistance of physicians to resort to
computerized aid for medical diagnosis
36. SOCIOECONOMIC ADVANTAGES
OF OUR PROGRAM
1. Reduces significantly the current numerous diagnostic errors and their dismal
consequences
2. Saves specialist consultations; important in emergencies and in underdeveloped
areas were specialists are not available
3. Reduces significantly the number of futile tests and procedures, by recommending
the best cost-benefit clinical data to investigate next in a patient. This saves
expense, risk, and discomfort for the patient and saves social medical resources
4. On the other hand, if best cost-benefit function proves probabilistically that a
diagnostic procedure is absolutely necessary, it will force managed care
organizations and insurers to authorize such procedures, otherwise curtailed for
being expensive
5. Facilitates and improves utilization reviews by hospitals, laboratories, and other
providers.
6. Protect health care providers against unfounded malpractice liability
7. Helps to overcome the current well known resistance of physicians to resort to
computerized aid for medical diagnosis
8. Some concepts of our diagnostic program are applicable and necessary to properly
implement the current renewed enthusiasm for long needed universal electronic
storage of medical records
38. DEMONSTRATION
A 32-year-old white female grade-school teacher with recent onset of
palpitations. Six months ago she lost her job, followed by
nervousness and depression. At that time a complete physical
examination and basic lab tests were normal; she was treated with an
antidepressant, without improving. Her medical condition gradually
worsened and recently the mentioned palpitations added to the clinical
picture, motivating the present consultation. The medical examination
revealed the following clinical data.
57. Abridged Data Cost Procedure Quantity
No Cost best cost-benefit clinical data: 0
Small Cost best cost-benefit clinical data: 0
Intermediate Cost best cost-benefit clinical data: 0
Great Cost best cost-benefit clinical data: 0
Total: 0
58. Abridged Data Cost Procedure Quantity
No Cost best cost-benefit clinical data: 35
G0001 MEDICAL HISTORY
C0819 (increased troponins with masked chest pain) G0008 IMAGING: CT AND/OR MRI
C0353 (abdominal pain, right lower quadrant)
C0829 (increased troponins without masked chest pain) C0834 (positron-emission tomography (PET) for pheochromocytoma)
C0800 (abdominal pain, left lower quadrant)
C0679 (plasma D-dimer ELISA) C0798 (mitral stenosis on cardiac magnetic resonance imaging)
C0781 (abdominal pain, left upper quadrant)
C0535 (ANA, antinuclear antibodies) C0682 (multislice spiral chest CT or MRI for pulmonary embolism)
C0780 (strong analgesics)
C0496 (latex agglutination test for rocky mountain spotted fever) C0612 (spiral CT scanning for pancreatic head carcinoma)
C0673 (pregnancy)
C0495 (indirect immunofluorescence assay for rocky mountain spotted C0785 (spiral CT scanning for pancreatic body or tail carcinoma)
C0401 (urinary tract symptoms, female)
C0045 (aortic dissection on MRI)
fever)
C0333 (malabsorption)
C0884 (sputum culture for pneumococcus)
C0296 (gross blood in stool)
C0395 (antigen detection in stool) G0009 SCANNING
C0262 (acholic, pale, clay-colored stool)
C0833 (nuclear medicine scintigraphy with MIBG for
C0384 (multiple blood cultures for acute endocardits)
C0215 (hypersensitivity to heat)
C0345 (serologic markers for Crohn's disease: ASCA) pheochromocytoma)
C0163 (anorexia, loss of appetite)
C0852 (serologic markers for ulcerative colitis: pANCA) C0233 (thyroid radionuclide scan)
C0127 (arthralgia)
C0244 (elevated plasma and urine catecholamines and metanephrines) C0665 (radionuclide imaging: viable myocardium, ventricular eject.)
C0004 (chest pain)
C0229 (suppressed TSH)
C0052 (history of angina pectoris)
C0228 (increased free, uncombined thyroxin, T four or T three) G0010 ENDOSCOPY
C0107 (diabetes)
C0151 (increased bilirubin) C0153 (colonoscopy and biopsy for Crohn's disease)
C0186 (colonoscopy and biopsy for ulcerative colitis)
C0131 (increased ESR)
G0002 PHYSICAL EXAMINATION
C0033 (leukocytosis)
C0794 (systolic thrill)
G0011 BIOPSY
C0678 (paradoxical bradycardia)
G0005 RADIOLOGY C0497 (biopsy and immunologic exam for R. rickettsii of a rash lesion)
C0562 (Horner's syndrome)
C0042 (abnormal aortic contour on chest x-ray; widening of C0399 (biopsy of small intestine for Giardia lamblia)
C0475 (chest dullness at percussion)
mediastinum)
C0379 (new/worsened regurgitant murmur)
C0085 (cardiac enlargement on imaging; usually due to right ventric. Great Cost best cost-benefit clinical data: 7
C0375 (meningism)
enlarg.)
C0365 (skin lesions)
C0086 (increased pulmonary artery and/or vascular engorgement on G0011 BIOPSY
C0337 (mouth lesions)
imaging)
C0332 (intestinal fistulas) C0611 (biopsy for pancreatic head carcinoma)
C0248 (jaundice, icterus) C0786 (biopsy for pancreatic body or tail carcinoma)
G0006 ULTRASOUND
C0238 (funduscopic abnormalities for pheochromocytoma)
C0044 (transthoracic aorta ultrasound for aortic dissection)
C0147 (tachypnea) G0013 ANGIOGRAPHY
C0015 (abdominal tenderness) C0047 (aortography)
G0007 ECHOCARDIOGRAM
C0018 (fever) C0071 (aortic valve stenosis on catheterization)
C0070 (aortic valve stenosis on echocardiogram)
C0019 (hypertension) C0090 (atrial septal defect on arteriography)
C0102 (mitral stenosis on echocardiogram)
C0022 (heart murmur) C0103 (mitral stenosis on catheterization)
C0054 (clinical data of advanced age) C0160 (pulmonary angiogram)
Intermediate Cost best cost-benefit clinical data: 16
C0055 (delayed second heart sound)
C0074 (right ventricular abnormal maximal impulse)
Total: 85
C0094 (mitral opening snap) G0005 RADIOLOGY
C0856 (upper GI series with small bowel follow-through for Crohn's
Small Cost best cost-benefit clinical data: 27 disease)
G0006 ULTRASOUND
G0004 LABORATORY TESTS
C0792 (transesophageal ultrasound for aortic dissection)
C0867 (nucleic acid urogenital probe tests for gonococcus)
C0788 (CA 19-9)
G0007 ECHOCARDIOGRAM
C0789 (CA 19-9)
C0797 (echocardiogram of atrial septal defect secundum)
C0705 (plasma ketones ++++)
59. Abridged Global Overview G0006 ULTRASOUND C0233 (thyroid radionuclide scan)
No Cost best cost-benefit clinical data: 35 C0044 (transthoracic aorta ultrasound for aortic dissection) D0011 (HYPERTHYROIDISM, GRAVES' DISEASE)
D0001 (AORTIC DISSECTION) total P: 818
G0001 MEDICAL HISTORY
total P: 169 present total P: 1000
C0353 (abdominal pain, right lower quadrant)
present total P: 1000 D0011 (HYPERTHYROIDISM, GRAVES' DISEASE)
D0021 (CROHN'S DISEASE, REGIONAL ENTERITIS)
G0007 ECHOCARDIOGRAM total P: 818
total P: 107
C0070 (aortic valve stenosis on echocardiogram) absent total P: 0
absent total P: 35
D0002 (AORTIC VALVE STENOSIS) G0010 ENDOSCOPY
C0781 (abdominal pain, left upper quadrant)
total P: 332 C0153 (colonoscopy and biopsy for Crohn's disease)
D0069 (PANCREATIC BODY-TAIL CARCINOMA)
present total P: 1000 D0021 (CROHN'S DISEASE, REGIONAL ENTERITIS)
total P: 176
D0002 (AORTIC VALVE STENOSIS) total P: 107
present total P: 800
total P: 332 absent total P: 0
D0069 (PANCREATIC BODY-TAIL CARCINOMA)
absent total P: 0 C0186 (colonoscopy and biopsy for ulcerative colitis)
total P: 176
D0022 (ULCERATIVE COLITIS)
Continues…
absent total P: 26
total P: 145
G0002 PHYSICAL EXAMINATION Intermediate Cost best cost-benefit clinical data:
present total P: 1000
C0794 (systolic thrill)
16 D0022 (ULCERATIVE COLITIS)
D0002 (AORTIC VALVE STENOSIS)
G0005 RADIOLOGY total P: 145
total P: 332
C0856 (upper GI series with small bowel follow-through for Crohn's absent total P: 0
present total P: 651
disease) G0011 BIOPSY
C0678 (paradoxical bradycardia)
D0021 (CROHN'S DISEASE, REGIONAL ENTERITIS) C0497 (biopsy and immunologic exam for R. rickettsii of a rash lesion)
D0007 (PULMONARY EMBOLISM)
total P: 107 D0035 (ROCKY MOUNTAIN SPOTTED FEVER)
total P: 157
present total P: 1000 total P: 272
present total P: 500
G0006 ULTRASOUND present total P: 1000
D0034 (PNEUMOCOCCAL PNEUMONIA)
C0792 (transesophageal ultrasound for aortic dissection) D0035 (ROCKY MOUNTAIN SPOTTED FEVER)
total P: 100
D0001 (AORTIC DISSECTION) total P: 272
present total P: 649
total P: 169 absent total P: 64
Continues…
present total P: 1000 C0399 (biopsy of small intestine for Giardia lamblia)
Small Cost best cost-benefit clinical data: 27 G0007 ECHOCARDIOGRAM D0026 (INTESTINAL GIARDIASIS)
G0004 LABORATORY TESTS C0797 (echocardiogram of atrial septal defect secundum) total P: 145
C0867 (nucleic acid urogenital probe tests for gonococcus) D0003 (ATRIAL SEPTAL DEFECT SECUNDUM) present total P: 1000
D0027 (DISSEMINATED GONOCOCCAL INFECTION) total P: 338 D0026 (INTESTINAL GIARDIASIS)
total P: 181 present total P: 1000 total P: 145
present total P: 1000 D0003 (ATRIAL SEPTAL DEFECT SECUNDUM) absent total P: 8
C0229 (suppressed TSH) total P: 338 Continues…
D0011 (HYPERTHYROIDISM, GRAVES' DISEASE) absent total P: 0
Great Cost best cost-benefit clinical data: 7
total P: 818 G0008 IMAGING: CT AND/OR MRI
C0834 (positron-emission tomography (PET) for pheochromocytoma) G0013 ANGIOGRAPHY
present total P: 1000
C0047 (aortography)
C0228 (increased free, uncombined thyroxin, T four or T three) D0012 (PHEOCHROMOCYTOMA)
D0001 (AORTIC DISSECTION)
D0011 (HYPERTHYROIDISM, GRAVES' DISEASE) total P: 364
total P: 169
total P: 818 present total P: 1000
present total P: 1000
absent total P: 0 C0798 (mitral stenosis on cardiac magnetic resonance imaging)
D0001 (AORTIC DISSECTION)
G0005 RADIOLOGY D0004 (MITRAL STENOSIS/RHEUMATIC HEART DISEASE)
total P: 169
C0042 (abnormal aortic contour on chest x-ray; widening of total P: 780
absent total P: 6
mediastinum) present total P: 1000
C0090 (atrial septal defect on arteriography)
D0001 (AORTIC DISSECTION) D0004 (MITRAL STENOSIS/RHEUMATIC HEART DISEASE)
D0003 (ATRIAL SEPTAL DEFECT SECUNDUM)
total P: 169 total P: 780
total P: 338
absent total P: 54 absent total P: 0
present total P: 1000
C0085 (cardiac enlargement on imaging; usually due to right ventric. G0009 SCANNING
D0003 (ATRIAL SEPTAL DEFECT SECUNDUM)
enlarg.) C0833 (nuclear medicine scintigraphy with MIBG for
total P: 338
D0003 (ATRIAL SEPTAL DEFECT SECUNDUM) pheochromocytoma)
absent total P: 0
total P: 338 D0012 (PHEOCHROMOCYTOMA)
Continues…
absent total P: 101 total P: 364
Total: 85
absent total P: 27
60. Abridged Data Cost Procedure Quantity
No Cost best cost-benefit clinical data: 2
G0002 PHYSICAL EXAMINATION
C0221 (thyroid bruit or thrill)
C0096 (rumbling diastolic murmur)
Small Cost best cost-benefit clinical data: 3
G0004 LABORATORY TESTS
C0229 (suppressed TSH)
C0228 (increased free, uncombined thyroxin, T four or T three)
G0007 ECHOCARDIOGRAM
C0102 (mitral stenosis on echocardiogram)
Intermediate Cost best cost-benefit clinical data: 2
G0008 IMAGING: CT AND/OR MRI
C0798 (mitral stenosis on cardiac magnetic resonance imaging)
G0009 SCANNING
C0233 (thyroid radionuclide scan)
Great Cost best cost-benefit clinical data: 1
G0013 ANGIOGRAPHY
C0103 (mitral stenosis on catheterization)
Total: 8
61. Abridged Global Overview
No Cost best cost-benefit clinical data: 2
Intermediate Cost best cost-benefit clinical data: 2
G0002 PHYSICAL EXAMINATION
G0008 IMAGING: CT AND/OR MRI
C0221 (thyroid bruit or thrill)
C0798 (mitral stenosis on cardiac magnetic resonance imaging)
D0011 (HYPERTHYROIDISM, GRAVES' DISEASE)
D0004 (MITRAL STENOSIS)
total P: 671
total P: 103
absent total P: 408
present total P: 1000
C0096 (rumbling diastolic murmur)
D0004 (MITRAL STENOSIS)
D0004 (MITRAL STENOSIS)
total P: 103
total P: 103
absent total P: 0
present total P: 176
G0009 SCANNING
Small Cost best cost-benefit clinical data: 3 C0233 (thyroid radionuclide scan)
G0004 LABORATORY TESTS D0011 (HYPERTHYROIDISM, GRAVES' DISEASE)
C0229 (suppressed TSH) total P: 671
present total P: 1000
D0011 (HYPERTHYROIDISM, GRAVES' DISEASE)
D0011 (HYPERTHYROIDISM, GRAVES' DISEASE)
total P: 671
total P: 671
present total P: 1000
absent total P: 0
D0011 (HYPERTHYROIDISM, GRAVES' DISEASE)
total P: 671 Great Cost best cost-benefit clinical data: 1
absent total P: 0 G0013 ANGIOGRAPHY
C0228 (increased free, uncombined thyroxin, T four or T three) C0103 (mitral stenosis on catheterization)
D0011 (HYPERTHYROIDISM, GRAVES' DISEASE) D0004 (MITRAL STENOSIS)
total P: 671 total P: 103
absent total P: 0 present total P: 1000
G0007 ECHOCARDIOGRAM D0004 (MITRAL STENOSIS)
C0102 (mitral stenosis on echocardiogram) total P: 103
D0004 (MITRAL STENOSIS) absent total P: 0
total P: 103
present total P: 1000 Total: 8
D0004 (MITRAL STENOSIS)
total P: 103
absent total P: 0
62. WHAT HAPPENED SO FAR ?
The theory behind our diagnostic system, its terminology, principles, and operation were
first published in 2006
63. COMPUTERIZED MEDICAL DIAGNOSIS:
A NOVEL SOLUTION TO AN OLD PROBLEM
INITIAL CONCLUSION OF DIAGNOSTIC QUEST?
CLINICAL DATA COLLECTION
AND SYNONYM HANDLING
CONCURRENT DIAGNOSES
CLINICAL DATA MATCHING
BEST COST-BENEFIT CLINICAL DATUM
NEXT TO INVESTIGATE
CREATION OF
CLINICAL DATUM LISTS
Step 1. Select clinical data not yet
investigated in the patient
CREATION OF
Step 2. Organize clinical data not yet
DIFFERENTIAL DIAGNOSIS LIST
investigated according to cost
category, diagnosis, PP value,
and S
CALCULATION OF PROBABILITY OF
DIAGNOSES. MINI-MAX PROCEDURE
Step 3. Recommend a new clinical
datum as best clinical datum
Step 1. Process
assuming it present
clinical data present
Step 4. Recommend a new clinical
Step 2. Process clinical data absent datum as best clinical datum
assuming it absent
Step 3. Create clinical data pairs
Step 4. Create clinical data pair MASKING
tables
Step 5. Calculate partial P that SIMULTANEOUS RECOMMENDATION OF
each clinical data pair SEVERAL BEST COST-BENEFIT
confers to each diagnosis CLINICAL DATA
Step 6. Create mini-max tables
SAFETY CHECK FOR RISK FLAGGED
DIAGNOSES AND CLINICAL DATA
Step 7. Determine total P
of the diagnosis
SAFETY CHECK FOR RELATED
CLINICAL ENTITIES
Step 8. Update
differential diagnosis list
CLINICAL PRESENTATION
Carlos Feder 2006
64. WHAT HAPPENED SO FAR ?
The theory behind our diagnostic system, its terminology, principles, and operation were
first published in 2006
A patent is pending
The computer program that runs our algorithm was written by Tomás Feder in 2007
This computer program provided a prototype, instrumental in confirming essentially all
our novel ideas, enabling to challenge real clinical cases of individual patients and the
theories behind, with excellent diagnostic results; it broadened our understanding of the
medical diagnostic process, dissipating earlier uncertainties, enabling the discovery of
new functions and eliciting ideas that opened an entire new realm in this field. So many
are the simplifications, additions, and codes that execute new functions, that they
justified writing a new book that we published in 2008
65. A PRACTICAL
COMPUTER
PROGRAM
THAT DIAGNOSES DISEASES
IN ACTUAL PATIENTS
Carlos Feder &
66. WHAT HAPPENED SO FAR ?
The theory behind our diagnostic system, its terminology, principles, and operation were
first published in 2006
A patent is pending
The computer program that runs our algorithm was written by Tomás Feder in 2007
This computer program provided a prototype, instrumental in confirming essentially all
our novel ideas, enabling to challenge real clinical cases of individual patients and the
theories behind, with excellent diagnostic results; it broadened our understanding of the
medical diagnostic process, dissipating earlier uncertainties, enabling the discovery of
new functions and eliciting ideas that opened an entire new realm in this field. So many
are the simplifications, additions, and codes that execute new functions, that they
justified writing a new book that we published in 2008
Our prototype includes so far 60 diseases in its database, and diagnoses beautifully real
patients afflicted by any of such diseases
67. WHAT NEXT?
Our diagnostic system, in order to accomplish all the envisioned benefits, must have its
database integrated with all currently known diseases, all their clinical data and
corresponding sensitivities. All complex clinical presentation models, which lists
related diagnoses must be created and also included in the database. This task requires
the cooperation of a team of medical specialists.
Our computer program is not a fancy program that opens colorful windows clicking on
menus or icons, but it satisfies all requisite functions of the algorithm and works
efficiently and accurately. This program would benefit from some cosmetic
improvement by an experienced professional programmer, able to make it more
glamorous and attractive.
We hope that Google or somebody else may help and support us with this project.
68. WHAT NEXT?
Our diagnostic system, in order to accomplish all the envisioned benefits, must have its
database integrated with all currently known diseases, all their clinical data and
corresponding sensitivities. All complex clinical presentation models, which lists
related diagnoses must be created and also included in the database. This task requires
the cooperation of a team of medical specialists.
Our computer program is not a fancy program that opens colorful windows clicking on
menus or icons, but it satisfies all requisite functions of the algorithm and works
efficiently and accurately. This program would benefit from some cosmetic
improvement by an experienced professional programmer, able to make it more
glamorous and attractive.
We hope that Google or somebody else may help and support us with this project.
69. WHAT NEXT?
Our diagnostic system, once implemented with all
known diseases and clinical data, is expected to
provide invaluable diagnostic benefits to patients,
physicians, nurses, health insurance companies,
malpractice lawyers, and the entire medical
establishment