Ben Hawkins: Week #6 (Thank You!)
My final week was a very different experience from the first. In part, because I changed locations, from the NICU to MICU (Medical Intensive Care Unit) where the patients were no longer children, but adults. The treatment of adult patients, at least in these cases, was much more complex, with a number of conditions overlapping. In one particular case, a woman had diabetes and kidney failure and other overlaying conditions that significantly complicated treatment. On the one hand, treatment for her kidney failure causes her insulin levels to fluctuate wildly, while keeping the insulin levels appropriate for her diabetes complicated her kidney failure. There were many more situations like this, where treatment decisions are not so easy as they were in the NICU or PICU. There the name of the game is simpler: keep the patient alive and breathing until they get better. But in many of these cases, with older patients the decision is not so clear cut.
In addition, I found the MICU to be a more stressful environment. Where a child might live or not (and all of them did, I'm happy to say), patients here are faced with the realization that they are going to die soon, and that there is nothing to be done. Another female patient was terminally ill with metastitized breast cancer, and treatment decisions were secondary to comfort. It was no longer a question of whether they could beat the condition or not, but what could be done to make her final days better. This was a sobering experience.
Outside of the MICU, I managed to spend some time in the laboratory. The hospital maintains its own set of labs for almost every imaginable test, and I spent the day with lab technicians, watching as they carried out urine protein alalysis or any number of other tests. The equipment they used was very sophisticated, and made light work of much of their procedures.
Beyond those experiences, much of my work was spent on the project. My project began with the previous student, who outlined the work to be done by future BME Immersion students. The project was to develop a lung model that could eventually be used in a predictive or diagnostic manner to analyze patient conditions who are on, or require mechanical ventilation. The development of the model follows and expands upon research already conducted and published in the field, but with an aim of making a more usable tool for doctors and clinicians in a bedside setting. By inputing some basic patient parameters; data gathered from examination, X-ray, and other imaging techniques, a model for the patient lung is constructed, based on an fine-grained lumped electrical parameter model. With this model in place, ventilator settings are input to the system and the model should predict the airway pressure and flow throughout the lung airway network. This information can be vital to doctors prescribing treatment to patients with distal airway damage, pneumothorax (collapsed lung), atelectasis, and any number of conditions requiring mechanical ventilation. For my part, I wrote the lung model itself in MATLAB. The patient parameter model and its interaction with the lung parameter model, as well as the ventilator model, will be left to future work. A basic set of parameters govern the operation of the model to this point.
To describe the model, we begin by condsidering the lumped parameter model of the lung. In this way the function of the airway is decomposed into three lumped electrical parameters: resistance (how much the airway resists the flow of air; largely a function of the cross-sectional area), compliance(how resistant the airway walls are to expansion and contraction), and inertance (the time required to start and stop the movement of air). The original lumped parameter model uses these values to model the whole lung. We expand this, and begin thinking about these three parameters as describing and descrete section of the airway (between bifurcations). The values for resistance, compliance, and inertance vary distally from the trachea, and can vary significantly in local sections due to various disease states. It is therefore important that the model be able to incorporate this sort of functionality. By allowing the parameter values to vary spatially over the lung, we accomplish this. By using an electric circuit model, we can solve for the potential and current (pressure and flow) using frequency domain analysis. This allows for a simplified solution, and by Fourier transform we can accomodate any periodic input waveform specified by the mechanical ventilator. The additional advantage to the modular design of the model allows independent development of each section, meaning the project can be undertaken by multiple people, in different locations, simultaneously.
To all those involved:
While participating in the program, I was attempting to absorb as much information as possible, but I wasn't being very effective at it. Most of what went on went right over my head, or in one ear and out the other. So many acronyms and so much jargon; it was overwhelming. But looking back now, even at what I wrote, I can observe my learning more easily retrospectively. Dr. Frayer told me at one point, that I wouldn't really understand how useful this experience was until much later, when my circumstances had changed. I now understand a little of that; I feel like I know a good deal more now than when I started, and about thing, and in ways, that I would never have access to otherwise. I would like to thank Dr. Steven Pon, in the PICU, for his patience and willingness to share his knowledge; to Dr. Joseph Schulman, for everything, but specifically answering all my questions and always willing to discuss conditions and treatment decisions. Also to Dr. Richard Lent, for allowing me to explore the laboratory spaces and spend time watching their operation; to Dr. Priscilla Winchester and Dr. David Trost, for allowing me to spend time in interventional radiology. To all these people, their collegues and assistants: Thank you so much for being an integral part of a truly unique program.
Thank you Rachel for organizing events and keeping everyone sane. Thank you Dr. Wang for having the vision to set this program in motion.
Most of all, I would like to thank Dr. William Frayer for his advice on everything from talking to doctors, recommending experiences, facilitating my experiencel; for helping me figure out where I needed to be, and where I wanted to go; for conversations and discussions; for help with my project; and most of all for his role as my mentor over the six week period.
Kindest regards,
Ben Hawkins
In addition, I found the MICU to be a more stressful environment. Where a child might live or not (and all of them did, I'm happy to say), patients here are faced with the realization that they are going to die soon, and that there is nothing to be done. Another female patient was terminally ill with metastitized breast cancer, and treatment decisions were secondary to comfort. It was no longer a question of whether they could beat the condition or not, but what could be done to make her final days better. This was a sobering experience.
Outside of the MICU, I managed to spend some time in the laboratory. The hospital maintains its own set of labs for almost every imaginable test, and I spent the day with lab technicians, watching as they carried out urine protein alalysis or any number of other tests. The equipment they used was very sophisticated, and made light work of much of their procedures.
Beyond those experiences, much of my work was spent on the project. My project began with the previous student, who outlined the work to be done by future BME Immersion students. The project was to develop a lung model that could eventually be used in a predictive or diagnostic manner to analyze patient conditions who are on, or require mechanical ventilation. The development of the model follows and expands upon research already conducted and published in the field, but with an aim of making a more usable tool for doctors and clinicians in a bedside setting. By inputing some basic patient parameters; data gathered from examination, X-ray, and other imaging techniques, a model for the patient lung is constructed, based on an fine-grained lumped electrical parameter model. With this model in place, ventilator settings are input to the system and the model should predict the airway pressure and flow throughout the lung airway network. This information can be vital to doctors prescribing treatment to patients with distal airway damage, pneumothorax (collapsed lung), atelectasis, and any number of conditions requiring mechanical ventilation. For my part, I wrote the lung model itself in MATLAB. The patient parameter model and its interaction with the lung parameter model, as well as the ventilator model, will be left to future work. A basic set of parameters govern the operation of the model to this point.
To describe the model, we begin by condsidering the lumped parameter model of the lung. In this way the function of the airway is decomposed into three lumped electrical parameters: resistance (how much the airway resists the flow of air; largely a function of the cross-sectional area), compliance(how resistant the airway walls are to expansion and contraction), and inertance (the time required to start and stop the movement of air). The original lumped parameter model uses these values to model the whole lung. We expand this, and begin thinking about these three parameters as describing and descrete section of the airway (between bifurcations). The values for resistance, compliance, and inertance vary distally from the trachea, and can vary significantly in local sections due to various disease states. It is therefore important that the model be able to incorporate this sort of functionality. By allowing the parameter values to vary spatially over the lung, we accomplish this. By using an electric circuit model, we can solve for the potential and current (pressure and flow) using frequency domain analysis. This allows for a simplified solution, and by Fourier transform we can accomodate any periodic input waveform specified by the mechanical ventilator. The additional advantage to the modular design of the model allows independent development of each section, meaning the project can be undertaken by multiple people, in different locations, simultaneously.
To all those involved:
While participating in the program, I was attempting to absorb as much information as possible, but I wasn't being very effective at it. Most of what went on went right over my head, or in one ear and out the other. So many acronyms and so much jargon; it was overwhelming. But looking back now, even at what I wrote, I can observe my learning more easily retrospectively. Dr. Frayer told me at one point, that I wouldn't really understand how useful this experience was until much later, when my circumstances had changed. I now understand a little of that; I feel like I know a good deal more now than when I started, and about thing, and in ways, that I would never have access to otherwise. I would like to thank Dr. Steven Pon, in the PICU, for his patience and willingness to share his knowledge; to Dr. Joseph Schulman, for everything, but specifically answering all my questions and always willing to discuss conditions and treatment decisions. Also to Dr. Richard Lent, for allowing me to explore the laboratory spaces and spend time watching their operation; to Dr. Priscilla Winchester and Dr. David Trost, for allowing me to spend time in interventional radiology. To all these people, their collegues and assistants: Thank you so much for being an integral part of a truly unique program.
Thank you Rachel for organizing events and keeping everyone sane. Thank you Dr. Wang for having the vision to set this program in motion.
Most of all, I would like to thank Dr. William Frayer for his advice on everything from talking to doctors, recommending experiences, facilitating my experiencel; for helping me figure out where I needed to be, and where I wanted to go; for conversations and discussions; for help with my project; and most of all for his role as my mentor over the six week period.
Kindest regards,
Ben Hawkins
posted by Ben Hawkins at 5:27 PM
0 comments
links to this post
