Method of and arrangement for optimizing disease diagnosis
First Claim
Patent Images
1. A method of optimizing the diagnosis of a condition of a patient tested, comprising the steps of:
- (a) acquiring electroencephalogram (analog signals) from the patient;
(b) mathematically determining a plurality of functions descriptive of the patient from the signals;
(c) establishing a set of indices for each function, each index having two states, each indicative of the condition of the patient;
(d) recognizing the state of each index for each function;
(e) generating an integrated pattern of the states of the indices from a plurality of the functions;
(f) storing a collection of index patterns, each containing a multitude of patterns of the states of indices for a multitude of patients whose condition is known;
(g) matching the generated integrated pattern against the stored collection of index patterns to obtain a plurality of raw scores;
(h) inputting the critical diagnostic factors, which effect the diagnosis of brain diseases into the device;
(i) establishing a set of weights for each of the critical diagnostic factors which effect the diagnosis of brain diseases, according to the importance of such factors;
(j) weighting the raw scores to obtain weighted scores; and
(k) presenting an optimized diagnosis based on the weighted scores.
0 Assignments
0 Petitions
Accused Products
Abstract
Disease is non-invasiveIy, accurately diagnosed at an early stage. A plurality of functions descriptive of the patient are mathematically determined. A set of indices for each function is established in advance. Each index has two states indicative of the patient'"'"'s condition. An integrated pattern of the states of the indices from a plurality of the functions is generated and matched against a stored collection of index patterns whose diagnosis is known. The diagnosis is optimized with the aid of weighting factors for such parameters as patient'"'"'s age and medical history.
-
Citations
2 Claims
-
1. A method of optimizing the diagnosis of a condition of a patient tested, comprising the steps of:
-
(a) acquiring electroencephalogram (analog signals) from the patient; (b) mathematically determining a plurality of functions descriptive of the patient from the signals; (c) establishing a set of indices for each function, each index having two states, each indicative of the condition of the patient; (d) recognizing the state of each index for each function; (e) generating an integrated pattern of the states of the indices from a plurality of the functions; (f) storing a collection of index patterns, each containing a multitude of patterns of the states of indices for a multitude of patients whose condition is known; (g) matching the generated integrated pattern against the stored collection of index patterns to obtain a plurality of raw scores; (h) inputting the critical diagnostic factors, which effect the diagnosis of brain diseases into the device; (i) establishing a set of weights for each of the critical diagnostic factors which effect the diagnosis of brain diseases, according to the importance of such factors; (j) weighting the raw scores to obtain weighted scores; and (k) presenting an optimized diagnosis based on the weighted scores.
-
-
2. An arrangement for optimizing the diagnosis of a condition of a patient tested, comprising:
-
(a) means for acquiring electroencephalogram (analog signals) from the patient; (b) means for mathematically determining a plurality of functions descriptive of the patent from the signals; (c) means for establishing a set of indices for each function, each index having two states, each indicative of the condition of the patient; (d) means for recognizing the state of each index for each function; (e) means for generating an integrated pattern of the states of the indices from a plurality of the functions; (f) means for storing a collection of index patterns, each containing a multitude of patterns of the states of indices for a multitude of patients whose condition is known; (g) means for matching the generated integrated pattern against the stored collection of index patterns to obtain a plurality of raw scores; (h) means for weighting the raw scores to obtain weighted scores; (i) means for presenting an optimized diagnosis based on the weighted scores; (j) said establishing indices including; (1) HW--Is the highest peak of the power spectrum in the delta band above a limit as well as too wide? (2) OH--Is the highest peak of the power spectrum in the theta band above a limit? (3) As---Is the main peak of the power spectrum of the alpha wave shifted to the theta band? (4) As+--Is the main peak of the power spectrum of the alpha wave shifted to the beta band? (5) EV--Is the transfer function curve too even in the low frequency range? (6) RC--Is the highest peak of the coherence in the theta band higher than that in the alpha band? (7) RT--Is the highest meek of the transfer function in the theta band higher than that in the alpha band? (8) PA--Does the shape of the phase angle wave resemble the letter "W"? (9) LF--Is there a late fluctuation of the impulse response? (10) L3--Is there a delay of the third peak of the impulse response? (11) F/A--Is the ratio of the occurrence frequency/amplitude of the amplitude histogram above a limit?
-
Specification