System for health monitoring on prosthetic and fixation devices

Abstract

A monitoring apparatus for a human body includes a node network with at least one motion sensor and at least one acoustic sensor. A processor is coupled to the node network, and receives motion information and acoustic information from the node network. The processor determines from the motion information and the acoustic information a source of acoustic emissions within the human body by analyzing the acoustic information in the time domain to identify an event envelope representing an acoustic event, determining a feature vector related to the event envelope, calculating a distance between the feature vector and each of a set of predetermined event silhouettes, and identifying one of the predetermined event silhouettes for which the distance is a minimum.

What is claimed is: 
     
         1 . A monitoring apparatus for a human body, comprising:
 a node network including:
 at least one motion sensor; and 
 at least one acoustic sensor; and 
   a processor coupled to the node network, configured to receive motion information and acoustic information from the node network and determine from the motion information and the acoustic information a source of acoustic emissions within the human body;   wherein the processor is configured to analyze the acoustic information in the time domain to identify an event envelope representing an acoustic event, determine a feature vector related to the event envelope, calculate a distance between the feature vector and each of a set of predetermined event silhouettes, and identify one of the predetermined event silhouettes for which the distance is a minimum.   
     
     
         2 . The monitoring apparatus of  claim 1 , further comprising an acquisition interface configured to sample signals from the at least one motion sensor and the at least one acoustic sensor and to provide the motion information and the acoustic information, wherein the acquisition interface is configured to omit sampling signals from the at least one acoustic sensor when the corresponding signal amplitude is below a threshold value related to the present noise amplitude. 
     
     
         3 . The monitoring apparatus of  claim 1 , further comprising a storage, wherein the motion information and the acoustic information is stored in the storage with time stamps. 
     
     
         4 . The monitoring apparatus of  claim 1 , wherein the acoustic information includes data representing an acoustic signal from one of the at least one acoustic sensor, the data normalized according to a maximum amplitude of the acoustic signal from the one acoustic sensor. 
     
     
         5 . The monitoring apparatus of  claim 1 , wherein the processor is further configured to fuse the motion information and the acoustic information. 
     
     
         6 . The monitoring apparatus of  claim 1 , wherein the apparatus is part of a prosthetic device. 
     
     
         7 . The monitoring apparatus of  claim 6 , wherein the prosthetic device is implantable in the human body. 
     
     
         8 . The monitoring apparatus of  claim 1 , wherein the apparatus is part of a fixation device. 
     
     
         9 . The monitoring apparatus of  claim 8 , wherein the fixation device is implantable in the human body. 
     
     
         10 . An non-transitory computer-readable medium comprising computer-executable instructions to:
 access time-stamped digital acoustic data and time-stamped digital motion data;   detect an acoustic event from the digital acoustic data;   calculate angle information from the digital motion data;   determine an angle-resolved event from the detected acoustic event and calculated angle information;   determine an event envelope and associated feature vector from the digital acoustic data corresponding to the detected acoustic event; and   identify a predetermined event silhouette with a least distance to the feature vector.   
     
     
         11 . The non-transitory computer-readable medium of  claim 10 , wherein the angle-resolved event is determined based in part on the time stamps of the digital acoustic data and the digital motion data. 
     
     
         12 . The non-transitory computer-readable medium of  claim 10 , wherein the predetermined event silhouette is one of a set of predetermined event silhouettes related to the angle-resolved event, and each predetermined event silhouette is determined in part using unbiased, hierarchical clustering. 
     
     
         13 . The non-transitory computer-readable medium of  claim 10 , further including instructions to select an event type based on the angle-resolved event. 
     
     
         14 . The non-transitory computer-readable medium of  claim 10 , further including instructions to identify a source of acoustic emissions in a human body from the identification of the predetermined event silhouette. 
     
     
         15 . The non-transitory computer-readable medium of  claim 14 , further including instructions to determine a health status of a portion of the human body from the event envelope. 
     
     
         16 . An apparatus, comprising:
 an acquisition interface configured to sample signals from at least one acoustic sensor, and to provide digitized acoustic data;   a storage including instructions; and   a processor;   wherein the instructions provide for configuring the processor to:
 detect an acoustic event from the digitized acoustic data; 
 determine an event envelope and corresponding feature vector for the acoustic event from the digitized acoustic data; 
 compare the feature vector to each of a set of predetermined silhouettes; and 
 identify one of the predetermined silhouettes with a least distance to the feature vector. 
   
     
     
         17 . The apparatus of  claim 16 , wherein the acquisition interface is further configured to sample signals from at least one motion sensor and to provide digitized motion data. 
     
     
         18 . The apparatus of  claim 17 , wherein the instructions further provide for configuring the processor to:
 calculate angle information from the digitized motion data;   determine an angle resolved event from the acoustic event and the angle information; and   select an event type based on the angle-resolved event.   
     
     
         19 . The apparatus of  claim 16 , wherein the instructions further provide for configuring the processor to create the set of predetermined silhouettes by configuring the processor to:
 determine an event envelope for each of a plurality of acoustic events;   calculate a feature vector for each event envelope;   identify clusters of feature vectors based on distance between feature vectors;   determine an event silhouette for each of a plurality of the identified clusters; and   group the event silhouettes into a set of predetermined silhouettes.   
     
     
         20 . The apparatus of  claim 19 , wherein the instructions further provide for configuring the processor to determine a classification metric based on a population of the clusters.