EARLY DIAGNOSIS OF DEMENTIA

• US 20090099848A1
• Filed: 10/16/2007
• Published: 04/16/2009
• Est. Priority Date: 10/16/2007
• Status: Abandoned Application

• Current Assignee: Keepalive Medical Ltd.
• Original Assignee: Keepalive Medical Ltd.
• Inventors: Lerner, Moshe, Bahar, Ofer
• Application Number:
  US11/873,019
• Publication Number:
  US 20090099848A1
• Time in Patent Office: Days
• US Class Current: 704/271
• CPC Class Codes: G10L 17/26 Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices

Abstract

The present invention is an innovative system and method for passive diagnosis of dementias. The disclosed invention enables early diagnosis of and assessments of the efficacy of medications for neural disorders which are characterized by progressive linguistic decline and circadian speech-rhythm disturbances. Clinical and psychometric indicators of dementias are automatically identified by longitudinal statistical measurements and track the nature of language change and/or patient audio features change using mathematical methods. According to embodiments of the present invention the disclosed system and method include multi-layer processing units wherein initial processing of the recorded audio data is processed in a local unit. Processed and required raw data is also transferred to a central unit which performs in-depth analysis of the audio data. The combined analysis enables the identification of the frequencies and nature of temporary relapses of linguistic decline, and provides essential data for diagnosing early stages of different types of dementias.

Claims

1. A method for the passive diagnosis of progressive linguistic decline and circadian speech-rhythm disturbances of a patient indicating dementia and neural disorders, said method comprising the step of:

continuous audio analysis in the premises of said patient over variable periods of time;

distinguishing the voice of said patient from all other recorded voices and noises;

storing audio data of said audio recording of said voice of the patient in clusters;

analyzing utterances in said clusters of said voice of said patient;

providing statistical information of at least one of the following;

utterances of the patient, speech patterns of the patient;

identifying symptomatic patterns in said clusters of said voice of said patient.

2. The method of claim 1 wherein said analysis of said utterances further includes analysis of speech patterns of the patient.

3. The method of claim 1 wherein said analysis is performed using a mathematic data processing model.

4. The method of claim 3 wherein said mathematic model is Hierarchical Hidden Markov Model (HHMM).

5. The method of claim 1 wherein said analysis includes at least one of the following:

utterance characteristics, phoneme characteristics, syllables characteristics, expressed emotion, speech characteristics.

6. The method of claim 5 wherein said utterance characteristics include at least one of the following:

utterance energy, utterance timing, utterance pitch.

7. The method of claim 1 wherein said speech characteristics includes identifying at least one of the following:

agrammatism, slow and labored speaking impaired articulation, literal paraphasias, neologisms, words finding pauses and hesitancy within speech, stuttered approximation of words, entropy of words, high-reoccurring coupling of words.

8. The method of claim 1 further including the steps of:

continuously monitoring biological parameters of said patient;

continuously identifying fluctuations in the circadian rhythms of said biological parameters of said patient.

9. The method of claim 8 wherein said biological parameters include at least one of the following:

body temperature, heartbeat rates, motion patterns.

10. The method of claim 1 further including the steps of:

continuously monitoring breathing patterns of said patient during sleeping hours;

continuously monitoring movement patterns of said patient during sleeping hours;

continuously identifying fluctuations in circadian rhythms of said patient in accordance with at least one of the following;

said fluctuations in breathing patterns, said fluctuations in movement patterns.

11. The method of claim 1 further including the step of identifying fluctuations in emotional speech in said clusters.

12. The method of claim 1 further including the step of identifying fluctuations in the fluency of predefined letters.

13. The method of claim 1 wherein said statistical information is analyzed in accordance with at least one of the following time periods:

seconds, minutes, hours, days, weeks, months, wherein said analysis is performed by comparing said utterances and said speech patterns in said different time periods.

14. The method of claim 1 further including the step of conducting active testing of said patient.

15. The method of claim 12 wherein said active testing includes at least one of the following:

voice testing, image testing.

16. A system for the passive diagnosis of progressive linguistic decline and circadian speech-rhythm disturbances of a patient indicating dementia and neural disorders, said system comprising:

at least one local unit located at the premises of said patient performing continuous audio recording in the premises of said patient over long periods of time and conducting preliminary analysis of said audio recording;

at least one central unit for gathering audio data from said local units, performing in-depth analysis of said audio recording and storing recorded and analyzed data;

a long-range data communication network for establishing periodic data transference of information between said local unit and said central unit.

17. The system of claim 16 further including:

at least one mobile carry-on unit for audio recording of said patient whenever said patient is outside the range of audio reception of said local unit;

a short-range communication network for establishing periodic data transference of information between said mobile unit and said local unit.

18. The system of claim 16 wherein said preliminary analysis includes filtering the voice of said patient from all other voices and noises.

19. The system of claim 16 wherein said local unit further includes a temporary memory unit for storing recorded and analyzed data for short periods of time.

20. The system of claim 16 wherein said analysis includes at least one of the following:

utterance characteristics, phoneme characteristics, syllables characteristics, expressed emotion, speech characteristics.

21. The system of claim 20 wherein said utterance characteristics include at least one of the following:

utterance energy, utterance timing, utterance pitch.

22. The system of claim 20 wherein said speech characteristics includes identifying at least one of the following:

agrammatism, slow and labored speaking impaired articulation, literal paraphasias, neologisms, words finding pauses and hesitancy within speak, stuttered approximation of words, entropy of words, high-reoccurring coupling of words.

23. The system of claim 16 further including a body temperature unit for continuously measuring fluctuations in the body temperature of said patient.

24. The system of claim 16 further including a breathing measuring unit for continuously measuring fluctuations in the breathing patterns of said patient.

25. The system of claim 16 further including a movement measuring unit for continuously measuring fluctuations in movement patterns of said patient.