INTRODUCTION

People learn differently. We can teach more effectively by knowing learning styles. We have to understand the learning styles of our students. In a group situation like classrooms and courses, we have to include various teaching methods to reach all learning styles.

There are three learning styles: auditory (listening, reading aloud), visual (seeing, reading) and kinesthetic (doing, writing, tactile).4,16 Auditory learners remember what they hear. They are good listeners, like verbal instructions and prefer to discuss new information. Visual learners remember what they read, prefer written instructions and like figures, pictures and diagrams. The kinesthetic learners have a strong need to manipulate things and use their hands. They like to be actively and physically involved and usually remember best through their experiences and experiments. Everybody has a preferred learning and teaching style. Knowing and understanding the students learning styles can help the instructors to structure their teaching more effectively.

To reach all three types of learners, an instructor must present the course material in multiple ways. Traditional lecture style is usually auditory. However in mathematics, the instructor writes everything on the board and the students take notes, and that is kinesthetic. The presentations can be considered both positive and negative. They give some visual input but may eliminate too much auditory. If the instructor asks the students to read the PowerPoint slides later and if the slides are the standard bullet point outline which does not stand alone later on, then the visual learning is limited. Very small-scale handouts discourage the kinesthetic learners by providing less room for them to take notes. Most engineers are kinesthetic learners. We, the bioengineering and biomedical engineering faculty, have to recognize all three learning styles and prepare our course content accordingly.

We have to continue enhancing our methods for electrophysiology education in classroom and laboratory by using technology and internet5,7 too. The new teaching methods we develop by using technology and cyber-environment including internet will improve our curriculum, textbooks and dissemination tools. As a result, scientists, researchers, and students from the fields of life sciences and engineering will benefit during their coursework, studies and lifelong learning experiences. The development of new hardware, middleware and software for cyber-environment allows the instructors and the students to access various teaching and learning resources. An interesting usage of technology is the development of software, tools and of resources for internet applications. Such software, tools and resources should provide user-friendly, easy-to-access and up-to-date education and scientific environment for the users; instructors and the students.

An interactive electrophysiology training resource was developed to present and to disseminate computational models of cellular bioelectric activity. Electrophysiology research projects have been integrated into education and training by this resource called iCell which is also known as the interactive cell modeling tool (http://ssd1.bme.memphis.edu/icell/) since 1998. iCell is used as simulation-based teaching and learning tool and even collaboration tool for electrophysiology. The site consists of JAVA-coded models of various cardiac cells and neurons, and provides simulation data of their bioelectric membrane activities at single cell level. Each JAVA-coded model is menu-driven and presents options to change model parameters or conditions, run and view simulation results. iCell simulations supplement cellular electrophysiology and transport knowledge students can gain from the physiology text-books.

The rationale for the development of iCell was:

  1. 1.

    to use technology in physiology education and training,

  2. 2.

    to provide a learning and teaching resource via internet for cellular electrophysiology,

  3. 3.

    to allow the user to understand the cellular physiology mechanisms, membrane transport and variations of action potentials and ion channels by running simulations interactively,

  4. 4.

    to provide a computer platform independent resource for JAVA-coded cellular models to be used and disseminated for teaching, learning and collaboration.; and

  5. 5.

    to supplement the students’ experiences of the physiology text book material with simulations which are the kinesthetic learning methods.

I prepare my course material and lectures by recognizing all three learning styles. I provide the course content in multiple ways and I understand that the bioengineering/biomedical engineering students have to be engaged. I mix my lecture delivery between visual presentations, mathematical derivations on the blackboard, exciting verbal information, open ended-questions and discussions, and simulations to help all students to connect to the course material. I add iCell and its simulations as teaching and learning modules to my courses with related physiology topics.

The primary goal of iCell was to develop and to disseminate JAVA-coded models of cellular activities, and to supplement the learning experiences of the students for the cellular physiology text-book material by allowing them to run simulations in iCell. The simulation-based learning is kinesthetic learning style and allows the student to be engaged in an interactive environment. iCell is also a supplemental teaching module for the instructor.

The main goal of this paper is to present the interactive cell modeling resource iCell and how it can be used as a simulation-based tool to enhance teaching and learning of the text-book material in cellular physiology.

BACKGROUND TO COMPUTATIONAL MODELING OF SINGLE CELLS AND ELECTROPHYSIOLOGY

After the first mathematical models of the cell membranes,6,11 a number of Hodgkin-Huxley type cell models have been developed to describe the electrophysiological behavior of single cells in the last four decades. These mathematical models have an electrical circuit representation for the cell membrane that is mainly described by passive and active transporters. These transporters are described as linear or nonlinear currents. The role and contribution of each passive and active transporter in different phases of the action potential still continues to be an important research topic to mathematical modelers. Especially the diastolic depolarization (pacemaker depolarization, Phase 4) in cardiac pacemaker cells, the repolarization in cardiac ventricular cells and the bursting phase in neurons are very complicated phases of action potentials since lots of ion currents are players with fast or slow time constants in these phases and the net ionic current can be balanced with many different combinations of the same set of currents. Thus the descriptions of these membrane ionic channel currents, pumps and exchangers have to be constrained by experimental data as the computational models are developed.

These computational models are developed to complement the physiological data with computational model development and simulation data, and to enhance the interdisciplinary field of computational bioengineering by integrating biological analysis and engineering analysis. The main research objectives of our computational models have been (1) to develop bio-physically detailed computational models based on experimental data, (2) to test several hypotheses related to control, disease & drug applied conditions, and (3) to expand our understanding of living systems’ physiology at the cellular level. Ultimately, the integration of knowledge from cell to bedside is desired. Quantitative approaches from molecular basis to clinical applications must be formulated since genes code ion channels, ion channels form the cell, the cells form the tissue and the tissues form the organ.

The major engineering techniques used in cellular model development are: circuit theory analysis, control theory analysis, numerical integration methods, phase plane analysis, perturbation analysis, parameter estimation, and statistics. Computer programming is utilized to develop and to code the mathematical equations to represent the electrophysiology behavior, to run simulations, and to analyze and visualize the model-generated data along with the experimental data. Based on the need and the complexity of the model system, computer platforms of high-end workstations or PCs, and different levels of computer languages can be used. The presentation of the simulation data (model-generated data) can be in various computer programming languages and scripts (Matlab, Unix, etc). The new dissemination methods of the computational models are being considered in addition to publishing mathematical equations in the journal publications.

IMPACT OF COMPUTATIONAL MODELING AT CELLULAR LEVEL

The following summarizes the impacts of the research in computational model development of the cellular bioelectric activity and the model-generated data in different disciplines of life sciences in general.

  1. 1.

    Biophysics and Physiology: The results of the computational studies expand our knowledge of the living systems at the cellular level in electrophysiology.

  2. 2.

    Clinical Physiology and Medicine: The insights gained and conclusions derived from the computational studies enhance our understanding of the biocomplexity of electrophysiology, and provide us with better knowledge to be used in the future in treatments for diseases in humans. We will also better understand the cells’ responses to various pathophysiological states with simulation data.

  3. 3.

    Pharmacology: The differences in membrane ionic currents, especially outward K+ currents in different species and cell types have very important practical implications. Different drugs are known to affect different ionic currents and to change action potential waveforms in different preparations under various conditions. A better understanding of the role of the ionic currents that control action potentials in different cells obtained from various species will provide motivation and explanations for species differences in treatment and drug actions, and also promote pharmacological research that may lead to the development of more specific drugs to be used in humans.

OVERVIEW OF THE INTERACTIVE CELL MODELING RESOURCE, iCELL

Since 1998, we have been disseminating our cellular computational models and other previously published models over the internet (http://ssd1.bme.memphis.edu/icell/) by coding them in JAVA. We have developed an interactive cell modeling tool (iCell) for electrophysiology that promotes simulation-based teaching, learning and collaboration. The tool integrates research and education for electrophysiology. The site consists of JAVA-coded models of various cardiac cells and neurons, and provides simulation data of their bioelectric membrane activities at single cell level. Each JAVA-coded model gives an overview of the cell, illustrates the cell membrane with an electrical equivalent circuit and cites the published modeling paper for further information. The cell models in iCell are grouped into versions, and cardiac or neuron “modelboxes.” Each JAVA-coded model allows the user to go through menu options to change model parameters, run and view simulation results. iCell also has a glossary section for the scientific terms to overcome the language problem among the scientists from different disciplines. The iCell site has been used by many scientists from the fields of biosciences, engineering, life sciences and medical sciences as a simulation-based teaching, learning and collaboration environment in addition to our usage in our biomedical engineering courses as a supplemental teaching module. The platform-independent software, iCell, provides an interactive and user-friendly teaching and learning web-resource, and also a training and collaboration environment for electrophysiology to be shared over the internet.

Development of iCell

Developing computational models in forms of mathematical equations take lots of time. The mathematical modeling experts have to spend dedicated time to describe the equations, and also to code, debug and compile the computer programs. Running simulations also requires extensive training and expertise. Computational model development of cellular bioelectric activity is an interdisciplinary field of bioengineering, which brings the scientists from engineering, biophysics and physiology with complementing skills for a cross-disciplinary collaboration and promotes their training opportunities. The developed computational models can be best understood by the users who have limited knowledge of model and equation development if they can run simulations interactively with the developed models. iCell meets the need to have a platform-independent, interactive, easy-to-use and user-friendly simulation-based teaching and learning resource, and also a training and collaboration environment for electrophysiology to be shared over the internet. The JAVA-coded cellular mathematical models in iCell are continuously updated and are free for non-profit usage in education.

Model Development: Java-Coded Cellular Models for Internet

The iCell models were presented as JAVA applets in html pages and can be executed in any JAVA-enabled browser. All of these JAVA applets were developed under JDK1.2 (Java Development Kit) of Sun Microsystems Inc. Each applet has a user-friendly interface and allows the user to go through menus to choose simulation protocols, change model parameters, run simulations, select and view the simulation data. An example of an html page including an applet model of the rabbit sinoatrial node cell is presented in Figs. 1 and 2.

FIGURE 1.
figure 1

A sample html page, containing a JAVAcoded model of the rabbit sinoatrial node cell,3 from iCell.

Full size image
FIGURE 2.
figure 2

A close-up of the html page in Fig. 1. JAVA-coded model of the rabbit sinoatrial node cell,3 from iCell.

Full size image

All of the JAVA applets have the similar structure. There are two classes in each of the JAVA applets. The first class is for the interface layout, GUI design and action handling; such as, where the figures will be graphed in the whole window, where the buttons are, and what should be done if the user pushes a button, etc. The second class is used to calculate the model results and to graph the area. All the ordinary differential equations (ODEs) are solved in the second class. The calculated data are stored in arrays and are graphed in the proper area. The second class will also handle “mouse down” action. When the user clicks on the figure in the window, the applet will display the exact coordinates that the mouse points to.

The Uniform Resource Locator (URL) technique is used in the applet for the Aplysia R15 bursting neuron. The simulations for this neuron can take the longest and the data saved may be very big too. Thus, this applet gives the user the option of running simulations or viewing the pre-calculated simulation results (Fig. 3). With the URL technique, the pre-calculated simulation data results were saved in our web server. If the user would like to view these results, then he chooses the simulation file to view and the pre-calculated results are displayed in the web browser instantly.

FIGURE 3.
figure 3

The JAVA-coded model of the R15 bursting neuron2 that can also display previously calculated results.

Full size image

The JAVA applets coded for the cellular models are grouped into versions, and also into cardiac and neuron “modelboxes.” The applets developed for the Cardiac Modelbox are for (1) a rabbit sinoatrial node cell model,3 (2) guinea pig ventricular cell model,10 (3) rabbit atrial cell model,9 (4) human atrial cell model,12 (5) a dog ventricular cell model,17 (6) a bullfrog atrial cell model,14 (7) a frog ventricular cell model15 and a rat ventricular cell.13 Currently, the applets in the Neuron Modelbox are for (1) a squid axon model6 and (2) an Aplysia R15 bursting neuron model.2

The JAVA applets coded for the cellular models are from our own published papers and from the published literature. During the internal development cycle, we validated the equations of the published model by running simulations and comparing the simulation results with the published simulation results of the paper. The authors of the paper were contacted when there were discrepancies with the published equations and the results, and their source codes were also requested. Before being placing in iCell, each JAVA-coded model was validated.

Teaching and Learning By iCell

We have used iCell as a supplemental educational tool for seven courses at the Joint Biomedical Engineering Program of University of Memphis and University of Tennessee. These courses are Life Sciences I for Biomedical Engineers (BIOM 7/8004 in 2001, 2002 and 2003), Life Sciences II for Biomedical Engineers (BIOM 7/8005 in 2004), Bioelectricity (BIOM 7/8203 in 2002), Medical Physics (PHY 4040 in 1998 and 2001), Computational Modeling of Cellular Systems (BIOM 809 in 1999), Physiological Control Systems (BIOM 7/8105 in 2001, 2003 and 2004) and Advanced Cardiac Electrophysiology (BIOM 817 in 2000). iCell was used in these courses when the lecture material was on the main topics of cellular membrane transport; and the subtopics included voltage-gated ion channels, resting membrane potential, action potential, drugs blocking ion channels, stimulus, threshold for various activities, different modes of action potentials, membrane permeability and conductance, membrane capacitance and related themes. As a result, the iCell simulations, that the students were asked to run, enhanced their knowledge of the textbook topics.

Simulation-Based Teaching by iCell

I have used iCell as a supplemental teaching tool when the course material covered the dynamics of cell membranes (e.g. action potential and the underlying ionic concentrations of calcium, potassium, sodium and chloride, and the ionic currents; ion channels, membrane pumps and exchangers). During my lectures, I run simulations with iCell to illustrate the electrical behavior of the cell membrane and the interactions between the ionic currents. I display simulations with different conditions to show the changes in the nonlinear behaviors of the cells. I also demonstrate to the students the significance of computational models and how the models can investigate more conditions than the experiments can since we can change the parameters and conditions numerically with exact values, and how the models can be used as predictive tools.

Simulation-Based Learning by iCell

The students in my courses used iCell as a self-learning tool while they were assigned to do homework with it. The assigned homework had certain simulation protocols to run. The students were asked to write a report and create tables and quantitative data of their simulation results for the changes they observed for the assigned simulation protocols. For example, Sample Assignments 1 and 2 (Appendix) are sections from the homework assigned to the students taking courses at the Joint Biomedical Engineering Program.

The aim of sample assignment 1 was for the students to learn about (1) the stimulus needed to create an action potential, (2) the changes in action potential characteristics due to blocking ionic currents, and (3) simulating one or multiple oscillating action potentials. This is a great simulation assignment to complement the students’ knowledge gained from the traditional cell biology textbook (Essential Cell Biology, [1]) of the topic “Membrane Transport” (Chapter 12). The students were introduced to nonlinear dynamics of the cellular bioelectric activity and associated mathematical equations by iCell assignments and simulations. Another aim of the exercise was to allow the students to develop their own virtual experimental protocols and open-ended questions. This aim contributes to their creativity and problem-based learning skills too.

The aim of sample assignment 2 was for the students (1) to simulate partial and complete blockades of ionic currents in various cardiac cell types, (2) to learn to analyze transient and steady-state simulation results of action potentials and ionic currents, and (3) to investigate the rate dependency of cardiac cells. It is not easy to block ion channels specifically in exact amounts in experiments, but such partial and specific, and also combined blocks can be simulated in computational models. Thus the simulation results of this assignment signify the important verification and prediction capabilities of the computer models to represent analyze and complement the physiological data. The model guides future experiments by predicting results for experiments that have not been performed. The prediction feature of the computational models allows the students to build trust for the computer simulations and their predictive capabilities.

All of these kinds of assignments provide a virtual experiment environment for the instructor to teach and for the students to learn by doing simulations. The simulation results provide both qualitative understanding and quantitative understanding of the cellular membrane transport activities and mechanisms, changes in ionic channels and action potentials, species and cell type differences for ionic mechanisms. The iCell simulations, that the students run, complement and enhance the knowledge that they gained from the textbook material.

The Assessment by the Students

Based on the student evaluations and assessments of their usage of iCell in the seven courses at the Joint Biomedical Engineering Program of University of Memphis and University of Tennessee, we concluded that most of the student users of iCell were not experts in developing, formulating and coding mathematical equations for the cellular models. However, they understood the iCell simulation results in a “qualitative” way and complemented their textbook knowledge. They also felt comfortable running simulations and analyzing the simulation data quantitatively by the tables that they created. Overall the students enjoyed doing “virtual” experiments by running simulations based on the iCell protocols that they were assigned to. They also liked designing their own virtual experiments and creating open-ended questions with iCell.

In our university system the students fill out the computerized Student Instructional Rating System (SIRS) forms for the courses they have taken. The students also write in a “free response” section. The information I have collected from those free response sections regarding the usage of iCell was that the students enjoyed the iCell exercises, found them informative, and appreciated their good connection and correlation with the course and text-book material. The students wrote that the instructor (Dr. Demir) did a good job of relating concepts learned in class to current research topics via these exercises which make her lectures interesting.

The Student Users

The students who are ready to take introductory level physiology courses can easily use iCell. The users of iCell were from the seven courses: Life Sciences I for Biomedical Engineers (BIOM 7/8004 in 2001, 2002 and 2003), Life Sciences II for Biomedical Engineers (BIOM 7/8005 in 2004), Bioelectricity (BIOM 7/8203 in 2002), Medical Physics (PHY 4040 in 1998 and 2001), Computational Modeling of Cellular Systems (BIOM 809 in 1999), Physiological Control Systems (BIOM 7/8105 in 2001, 2003 and 2004) and Advanced Cardiac Electrophysiology (BIOM 817 in 2000), at the Joint Biomedical Engineering Program of University of Memphis and University of Tennessee. 85% students were Biomedical Engineering (BME) graduate students, 5% were BME undergraduate students and 10% were non-BME students (from the nursing school, speech and audilogy school, MD program). Only 2% of the students had knowledge in computational modeling in bioelectricity before taking these courses.

Simulation-Based Teaching, Learning and Collaboration Environment

This modeling tool iCell was also used as a collaboration site among our physiology and biophysics colleagues interested in simulations of cell membrane activities. We collaborate with physiology and biophysics scientists for our own model development projects. These physiology and biophysics colleagues do not need to know the details of mathematics or the formulations in the models, but they can run simulations on their own in iCell and get a preliminary understanding of the simulations results before we start developing models and research discussions and meetings.

Moreover, the iCell site has been used by many scientists and students (http://ssd1.bme.memphis.edu/icell/people.htm) from the fields of biosciences, engineering, life sciences and medical sciences in Argentina, Belgium, Brazil, Canada, China, England, Germany, Greece, Ireland, Japan, Korea, the Netherlands, New Zealand, Spain, Taiwan, Turkey and USA as a simulation-based teaching, learning and collaboration environment. Some of those professors who requested permission to use iCell in their courses are from the following listed universities and disciplines: Johns Hopkins University (Biomedical Engineering, Medical School), Medical University of South Carolina (Cardiology) University of Utah (Bioengineering, Medical School), North Carolina State University (Biological and Agricultural Engineering), Southampton University United Kingdom, University of Leeds (Biomedical Sciences), Georgia Institute of Technology (Electrical Engineering), University College Dublin (Veterinary Physiology and Biochemistry), Texas A & M University (Bioengineering, Physiology and Pharmacology), University of Georgia (Biological Agricultural Engineering) University of Washington (Bioengineering) California State Univ. Fullerton (Biological Sciences), Hope College (Biology), Kansas University Medical Center (Molecular and Integrative Physiology), Georgia University (Anatomy and Physiology, Nursing Anesthesia), University of Amsterdam (Cardiology, Physiology), University Hospital Groningen (Cardiology), University of Utrecht (Physiology), and Universitat de les Illes Balears (Animal Physiology).

DISCUSSIONS

The simulations provided by iCell and other resources such as CellML (http://www.cellml.org/public/news/index.html, an XML based depository of models), Virtual Cell (http://www.nrcam.uchc.edu/, a general framework for the spatial modeling and simulation of cellular physiology), JSIM (http://nsr.bioeng.washington.edu/PLN/Software/, Java-based simulation system for building and analyzing quantitative numeric models.), simBio Cell/Biodynamics Simulation Project of Kyoto University (http://www.biosim.med.kyoto-u.ac.jp/e/index.html), E-Cell (www.e-cell.org, modeling biochemical and generic processes), CMISS (continuum mechanics, image analysis, signal processing and system identification) and Gepasi (http://www.gepasi.org, software package for modeling biochemical systems) will continue signifying the important verification and prediction capabilities of the computer models to represent, analyze and complement the physiological data and knowledge. The review by Hunter and Borg8 entitled “Innovation: Integration from proteins to organs: the Physiome Project” provides a comprehensive comparison of different resources.

The model development demonstrates that computational models have to be constructed from experimental data, not only to explain and to verify the data that they are based on, but also to predict results for experiments that have not been performed and to guide future experiments. The verification and prediction features of the computational models will allow scientists to build trust for the computer simulations and their predictive capabilities. The utilization of simulation protocols demonstrates the significance of the mathematical models and their importance in determining “computational hypothesis,” and that the computational insights generated will enhance knowledge in molecular biology, biophysics and physiology. The dissemination of computational models and their simulators via interactive web sites contributes to simulation-based learning, teaching and collaboration significantly can supplement learning and teaching in a kinesthetic way.

Overall, computational modeling and simulation results continue to advance our understanding of living systems at the cellular level in cardiac electrophysiology while promoting collaborations and training in interdisciplinary field of bioengineering between scientists in life scientists and engineers. The presentation of computational models in user friendly, interactive and menu driven software are important in bringing collaborators of different disciplines and training scientists and students in cross-disciplinary projects.

LIMITATIONS AND FUTURE WORK

iCell and its presentation in this paper have limitations. However these limitations provide the basis and the motivation for future work.

  1. 1.

    iCell does not provide a platform for user to develop formulas and computational models for cellular activity. iCell allows the users to run simulations of the previously developed cellular models by changing parameters. The users view the calculated results (variables of the models). A major virtue of iCell is its simulator environment which is the convenient and model-specific interface. We, the developers, continue to develop and add more models to iCell even though the users cannot add models.

  2. 2.

    This paper does not focus on assessing the efficacy of using iCell in teaching. The primary goal of this project was to develop and provide a simulation-based teaching module to supplement text-book knowledge. The usage of our students has been tracked for assessment purposes if the students wanted to provide any feedback. Two sections “The Assessment by the Students” and “The Student Users” provided this preliminary assessment. Future work can assess the learning outcomes of groups that use iCell and of groups that do not use iCell for the same text book material. Another study can be on comparing the usage of iCell with other similar modeling and simulator resources. Another assessment can explore the outcomes for those who use iCell for distance learning, and those for classroom learning. There was not any report of iCell performance problem by concurrent users. A future project can study the usage and performance of iCell by excessive concurrent users.

CONCLUSIONS

We are enhancing iCell by developing more JAVA-coded models for the Cardiac and Neuron Modelboxes. We are also adding more features to analyze the simulation results. The platform-independent software resource, iCell, provides us with an interactive and user-friendly teaching, learning and collaboration environment for electrophysiology to be shared over the Internet and to supplement the text-book material with simulations. iCell will continue to motivate the development of the students and scientists in the cross-disciplinary fields of engineering and life sciences in different parts of the world. The usage of simulations for teaching and learning will also advance simulation-based engineering and sciences for research and development with kinesthetic methods.

In short, the development and usage of iCell promotes the usage of simulation-based learning and teaching for cellular electrophysiology education, supplements text-book material, allows the user to better understand the cellular membrane transport mechanisms by running simulations, and provides a computer platform independent resource for cellular models to be shared for teaching, learning and research collaboration over the internet.