System and method for detecting and measuring the condition of intraluminal esophageal mucosa
Disclosed are catheter systems and methods of using the catheter system to acquire mucosal impedance data of a patient. Also disclosed are methods of classifying or otherwise identifying esophageal conditions in a subject based on mucosal impedance data acquired using a catheter system. Unlike conventions systems that require subjective input from a physician to render a diagnosis, the systems and methods described herein can utilize the mucosal impedance measurements to generate a probability that the subject's esophagus corresponds to an esophageal condition or a set of esophageal conditions. In one embodiment, a classification model is used to generate the probability. The classification model may generate the probability based at least in part on a change in the mucosal impedance measurements between a distal location and a proximal location on the subject's esophagus.
1. A catheter system comprising:
a catheter including an elongate shaft extending between a proximal end and a distal end;
a plurality of impedance sensing electrodes configured between the proximal end and the distal end;
an inflatable and deflatable balloon configured to the elongate shaft;
a computer system communicatively coupled to the catheter, and wherein the computer system is programmed to:
control the inflatable and deflatable balloon to expand to place at least a portion of the impedance sensing electrodes in contact with an esophageal mucosa on the interior esophageal wall of a subject;
acquire mucosal impedance data of the esophageal mucosa at a plurality of axial positions between the distal position and the proximal position of the catheter by directing an electric current through the esophageal mucosa via the impedance sensing electrodes;
generate a report that includes a probability that at least a portion of the subject's esophageal mucosa corresponds to one or more esophageal condition, wherein the probability is generated based at least in part on the mucosal impedance data acquired between the proximal end and the distal end of the catheter, and
wherein the probability is generated using a classification model that includes a multinomial logistic regression, wherein the multinomial logistic regression generates the probability based at least in part on fitting the mucosal impedance data at two or more of the axial positions to generate a classification slope and a classification intercept, and generating the probability using the classification slope and the classification intercept.
2. The catheter system of claim 1 , wherein the computer system is further programmed to:
generate a probability that the esophageal mucosa corresponds to a set of esophageal conditions, wherein the probability is generated based at least in part on the mucosal impedance data acquired between the proximal end and the distal end of the catheter, wherein the set of esophageal conditions includes gastroesophageal reflux disease (GERD), eosinophilic esophagitis (EoE), and non-gastroesophageal reflux disease (NERD).
3. The catheter system of claim 1 , wherein an exterior surface of the catheter includes at least two impedance sensing electrodes at a plurality of axial positions along the length of the catheter.
4. The catheter system of claim 1 , wherein the computer system is further programmed to:
identify and remove non-contact impedance data from the acquired mucosal impedance data to generate corrected mucosal impedance data, wherein the non-contact impedance data corresponds to impedance data generated when the impedance sensing electrodes are in contact with a remnant in the subject's esophagus that is different than the esophageal mucosa, and wherein the computer system identifies and removes the non-contact impedance data if the data corresponds to one or more of the following conditions:
(a.) acquired impedance data that exceeds a threshold impedance measurement, wherein the threshold measurement is 5000 Ohms; and
(b.) acquired impedance data that has a studentized residual above 3.
5. The catheter system of claim 1 , wherein the computer system is further programmed to:
store an individual correction factor for at least a portion of the impedance sensing electrodes in a memory of the computer system; and
generate corrected mucosal impedance data by applying the individual correction factor to the mucosal impedance data acquired by the corresponding impedance sensing electrode.
6. The catheter system of claim 5 , wherein the computer system is further programmed to:
generate the individual correction factors by calibrating at least a portion of the impedance sensing electrodes using a saline solution having a known impedance and concentration.
7. A catheter system comprising:
a catheter including an elongate shaft extending between a proximal end and a distal end;
a plurality of impedance sensing electrodes configured between the proximal end and the distal end;
a computer system communicatively coupled to the catheter, and wherein the computer system is programmed to:
acquire mucosal impedance data of esophageal mucosa on the interior esophageal wall of a subject using impedance sensing electrodes at a plurality of axial positions between the distal end and the proximal end of the catheter by directing an electric current through the esophageal mucosa via the impedance sensing electrodes;
generate a report that includes a probability that at least a portion of the subject's esophageal mucosa corresponds to one or more esophageal condition, wherein the probability is generated based at least in part on the mucosal impedance data acquired between the proximal end and the distal end of the catheter, and
wherein the probability is generated using a classification model that includes a multinomial logistic regression, wherein the multinomial logistic regression generates the probability based at least in part on fitting the mucosal impedance data at two or more of the axial positions to generate a classification slope and a classification intercept, and generating the probability using the classification slope and the classification intercept.
8. The catheter system of claim 7 , wherein the computer system is further programmed to:
generate a probability that the esophageal mucosa corresponds to a set of esophageal conditions, wherein the probability is generated based at least in part on the mucosal impedance data acquired between the proximal end and the distal end of the catheter, wherein the set of esophageal conditions includes gastroesophageal reflux disease (GERD), eosinophilic esophagitis (EoE), and non-gastroesophageal reflux disease (NERD).
9. The catheter system of claim 7 , wherein an exterior surface of the catheter includes at least two impedance sensing electrodes at a plurality of axial positions along the length of the catheter.
10. The catheter system of claim 7 , wherein the computer system is further programmed to:
identify and remove non-contact impedance data from the acquired mucosal impedance data to generate corrected mucosal impedance data, wherein the non-contact impedance data corresponds to impedance data generated when the impedance sensing electrodes are in contact with a remnant in the subject's esophagus that is different than the esophageal mucosa, and wherein the computer system identifies and removes the non-contact impedance data if the data corresponds to one or more of the following conditions:
(a.) acquired impedance data that exceeds a threshold impedance measurement, wherein the threshold measurement is 5000 Ohms; and
(b.) acquired impedance data that has a studentized residual above 3.
11. The catheter system of claim 7 , wherein the computer system is further programmed to:
store an individual correction factor for at least a portion of the impedance sensing electrodes in a memory of the computer system; and
generate corrected mucosal impedance data by applying the individual correction factor to the mucosal impedance data acquired by the corresponding impedance sensing electrode.
12. The catheter system of claim 11 , wherein the computer system is further programmed to:
generate the individual correction factors by calibrating at least a portion of the impedance sensing electrodes using a saline solution having a known impedance and concentration.