What is claimed is:
1. A voice recordingreproducing device implemented by using an adaptive differential pulse code modulation method comprising:
an analog-digital converter to sample analog voice signals and to convert them to digital voice data;
recording means for quantizing a difference in the sampled data value between a digital voice data and a subsequent and neighboring digital voice data obtained by said analog-digital converter in accordance with the width for quantizing corresponding to the difference, for obtaining encoded data by encoding the quantized difference and then for splitting the encoded voice data into voice blocks each having a predetermined size;
reproducing means for reproducing digital voice data, at a time of a fast-forward reproduction, by sequentially decoding specified data composed of the encoded data contained in the voice block to be reproduced and obtained by removing and culling a part of voice blocks from a series of the split voice blocks;
a digital-analog converter to convert reproduced digital voice data to analog voice signals, and
whereby said recording means is operated to add just previously sampled data and a parameter used to decide said width for quantizing to a head of said encoded data of each voice block and said reproducing means is operated, at the time of the fast-forward reproduction, to decode said head coded data of the voice block to be reproduced based on said additional data.
2. The voice recordingreproducing device according to claim 1, wherein said data constituting said difference contains data of a multiple of said width for quantizing and of polarity causing an increase or decrease of said difference.
3. The voice recordingreproducing device according to claim 1, wherein said encoded data contains data of the multiple of said width for quantizing and wherein said reproducing means is operated to decode said head digital voice data of each voice block by using data of the multiple of said width for quantizing contained in said encoded data, just previously sampled data and said parameter used to decide said width for quantizing.
4. The voice recordingreproducing device according to claim 1, wherein said recording means is provided with an ADPCM (Adaptive Differential Pulse Code Modulation) analyzer.
5. The voice recordingreproducing device according to claim 1, wherein said reproducing means is provided with an ADPCM synthesizer.
6. The voice recordingreproducing device according to claim 1, wherein a format of said encoded data is composed of a voice start address region, a voice end address region and an ADPCM data region to its head of which said additional data are added.
7. The voice recordingreproducing device according to claim 1, wherein said format of said encoded data further includes a sampling frequency region.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
1. A method for controlling a medical diagnostic ultrasound imaging system, the method comprising:
obtaining an ultrasound image;
identifying, with a processor and trained classifier, an anatomical feature in the image;
setting, with the processor, an imaging value as a function of the anatomical feature; and
subsequently imaging with the medical diagnostic ultrasound system, the imaging responsive to the imaging value set as the function of the anatomical feature.
2. The method of claim 1 wherein identifying the anatomical feature comprises extracting the anatomical feature with a multi-class classifier.
3. The method of claim 2 wherein extracting comprises applying the multi-class classifier with feature vectors selected using a boosting framework.
4. The method of claim 2 wherein extracting comprises applying the multi-class classifier with a tree and cascade structure.
5. The method of claim 1 wherein identifying comprises identifying a canonical view.
6. The method of claim 1 wherein setting comprises setting a beam location for a color pan box, a Doppler gate or both.
7. The method of claim 1 wherein setting comprises setting a B-mode display depth, a B-mode frequency, or both.
8. The method of claim 1 wherein identifying and setting are performed during an imaging session substantially in real-time, and wherein subsequently imaging is performed during the imaging session.
9. In a computer readable storage medium having stored therein data representing instructions executable by a programmed processor for controlling a medical imaging system, the storage medium comprising instructions for:
deriving anatomical information from a medical image with machine-learned feature vectors; and
setting at least one imaging parameter of the medical imaging system as a function of the anatomical information.
10. The instructions of claim 9 wherein deriving the anatomical information comprises extracting the anatomical information with a multi-class classifier.
11. The instructions of claim 10 wherein extracting comprises applying the multi-class classifier with feature vectors selected using a boosting framework.
12. The instructions of claim 10 wherein extracting comprises applying the multi-class classifier with a tree and cascade structure.
13. The instructions of claim 9 wherein deriving comprises deriving the anatomical information from a medical diagnostic ultrasound image, and wherein setting comprises setting the at least one imaging parameter of a medical diagnostic ultrasound imaging system.
14. The instructions of claim 9 wherein deriving comprises identifying cardiac structure associated with a particular view of a heart.
15. The instructions of claim 9 wherein deriving comprises identifying a heart valve, and wherein setting comprises positioning a color pan box, a Doppler gate or both for subsequent images.
16. The instructions of claim 9 wherein deriving and setting are performed during an imaging session substantially in real-time, images obtained in the imaging session responsive to the at least one imaging parameter set as a function of the anatomical information.
17. The instructions of claim 9 wherein setting comprises setting a B-mode display depth, a B-mode frequency, or both as a function of the anatomical information.
18. A medical diagnostic imaging system, the imaging system comprising:
a memory operable to store a medical image; and
a processor operable to apply a classifier based on machine learning to the medical image, the classifier operable to identify a canonical view, tissue structure, flow pattern or combinations thereof from the medical image, the processor operable to control imaging as a function of the view, structure, pattern or combination thereof.
19. The imaging system of claim 18 wherein the classifier comprises a multi-class boosting classifier with a tree and cascade structure.
20. The imaging system of claim 18 wherein the processor comprises medical diagnostic ultrasound imaging system processor, the medical image comprises an ultrasound image, and the processor being operable to control comprises positioning a color pan box, positioning a Doppler gate, setting a B-mode display depth, setting a B-mode frequency, or combinations thereof.
21. The imaging system of claim 18 wherein the classifier is operable to identify cardiac structure associated with a particular view of a heart.
22. A method for identifying information from a medical image, the method comprising:
applying a multi-class classifier to the medical image; and
identifying a particular canonical view, tissue structure, flow pattern or combinations thereof from the medical image as a function of applying the multi-class classifier.
23. The method of claim 22 wherein applying the multi-class classifier comprises applying the multi-class classifier with feature vectors learned from boosting.
24. The method of claim 22 wherein applying the multi-class classifier comprises applying the classifier with a tree and cascade structure.
25. The method of claim 22 further comprising:
automatically setting imaging parameters of an imaging system as a function of the particular canonical view, tissue structure, flow pattern or combination thereof.
26. In a computer readable storage medium having stored therein data representing instructions executable by a programmed processor for identifying information from a medical image, the storage medium comprising instructions for:
applying a multi-class classifier to the medical image; and
identifying a particular canonical view, tissue structure, flow pattern or combinations thereof from the medical image as a function of applying the multi-class classifier.