Friday, January 29, 2010

Notre Dame CBE News: Stinson - Remick Hall is open! Thanks!

Notre Dame CBE news!  Stinson-Remick Hall is open!  Thanks to all who contributed!
I am pleased to announce that as of this semester, Stinson-Remick Hall, a new teaching and research building for the engineering college has opened.  It is built around a clean room and 2 floors of multiple function rooms that comprise the McCourtney Learning Center -- which is used by undergraduate engineers of all disciplines and class levels.  Research activities include new energy technologies that employ ionic liquids, nano-electronics and nano-optoelectronics for the next generation of electronic devices, micro and nano fluidic devices for medical diagnostics and therapeutics, nanoscale bio-electronics to address problems at the interface of living tissue and artificial devices and actinide chemistry technologies to address nuclear waste.
Faculty and graduate students from the Chemical and Biomolecular Engineering Department are playing an active role in most of these research efforts.  Our undergraduates … those so lucky as to be taking “Junior Lab” are enjoying the brand new laboratory built just for this purpose.  Rather than toiling for hours below ground, they now have a lab with windows (well into the atrium) looking out from its prominent position at the top of the main staircase.  “Senior Lab” will use this room as well.  Our biomolecular engineering laboratory will be housed in another new lab just across the atrium.
This new building will enable Notre Dame Engineering to grow to a new, higher level of prominence.  The better laboratories and the clean room in the new facility will enable research that could not be done previously at Notre Dame and will attract talented students and faculty who might not have otherwise come to Notre Dame.  For undergraduates, in addition to more research opportunities and better chemical engineering laboratories, the ample study and group-work space will make all of those late nights not only more pleasant, but more productive.       
So on behalf of the chemical engineering students and faculty, I would like to all of you who contributed your hard-earned money to this building!  Further we want to laud the efforts of Professor Frank Incropera who was Dean of Engineering while planning and fundraising was completed for this building.  Of course, none of this would have been possible without the support of the central administration of the University.















Friday, January 22, 2010

Random Thought: Mass balances and drug dosing

It happened again. I have mentioned many times in the mass and energy balance class that I could not understand how drug dosing is done.   I suggested that this needed attention by the medical profession (as brazen as this seems!)    That is,  the adult dose for almost all drugs, is the same -- be it one, two, three or 4 times per day.  I could not help but wonder how small women and really large men would need the same dose when a simple mass balance tells us that if it some systemic concentration of drug is needed for efficacy, then the dose should scale roughly as weight.  If there is partitioning of the compound in different types of tissue (e.g., fat tissue which would be hydrophobic), then perhaps a more nuanced criterion is necessary.  However, in either case, all adults are not equal (despite the claim in the founding documents of the US!)

Well, as reported by the New York Times (http://www.nytimes.com/2010/01/20/health/26regimens.html?ref=health), a new paper in Lancet (http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(09)60743-1/fulltext ), has suggested the need to tailor doses of antibiotics to a person’s size.  They note that obese patients may not get enough drug to clear an infection and that underdosing could be contributing to antibiotic resistance of various bacteria.  There is an accompanying editorial (http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60073-6/fulltext), which I can also not see (##!!) that suggests the need to study this issue with careful trials.

Beyond the irresistible “duh”, this is actually a very interesting issue for chemical engineers.  The first is that there are good reasons to have FDA approved dosing rules.  You would not want physicians to be “winging it” for every patient -- particularly if they could not solve the transient mass balance equation with simple elimination and reaction terms.  Further, it is critical that patients take the correct dosing.  So three small pills and one big pill, taken every 3 hours is probably problematic for most people (except engineers -- we can count and if you are a student you don’t sleep!).  However, the question of if a particular condition willrespond better to a constant concentration or peaks and valleys would seem not to have a general answer.  Maybe two big doses, is better than four doses of 1/2 the size.  

Looking a bit further afield, we could consider how medicine is likely to progress and how chemical engineers will be involved.  Various “artificial pancreas” devices are in different stages of development.  Chemical engineering plays an essential role in glucose sensing, determination of the “control scheme” and materials necessary for construction.  The implantable drug delivery system, either permanent or temporary, idea could be extended to many more diseases with the essential advantage being feedback control.  Either the drug level could be controlled or the dose could be altered in response to levels of something that is a response to the disease such as a specific cytokine or a blood toxin.  

Cochlear implants have become a common and successful treatment to restore hearing.  Soon there will be vision devices.  This technology, which allows linking of electronic and electromechanical devices, to nerves and other human tissue, could allow blood pressure monitoring, heart rate analysis, or other medical measures that determine health of the patient on a continuous basis.  We might also envision “health stations” for minor ailments where various sensors are used to check out patients and provide a first level screening that helps decide if a physician is needed.  The physician counseling could be remove -- although this invokes a vision of a “call center” for computer software …. -- and the consultations could be with the next available person from a large pool rather than the one overworked Doc at MedPoint.  This approach may be distasteful to some, but potentially could save a lot of money with out limiting eventual use of the best treatments.   

All of these technologies will require the skills of chemical engineers and there are certain questions that ONLY chemical engineers can answer.  Thus chemical engineers will play a key role in the future of health care both to enable treatments that are not yet possible and to save money to allow for care of many more people. 

Monday, January 18, 2010

Random Thought: Statistics versus mechanistics

Most likely you have seen a TV commercial for one of the “statin” drugs, say Lipitor  or Crestor, which are used to reduce “bad cholesterol” (LDL) and raise “good cholesterol” (HDL).  Presumably these do, in fact, cause these level changes for some high fraction of patients ( … who do not have liver disease, who are not nursing, who are not pregnant nor might become pregnant!).

Also, presumably the medical studies that show a link between above median levels of LDL and heart disease are valid.

Why is it then that the drug description sheet (for Crestor, from the manufacturer: http://multivu.prnewswire.com/mnr/astrazeneca/41126/docs/41126-crestor_differentiators_fact_sheet.pdf) states that: “CRESTOR has not been determined to prevent heart disease, heart attacks or strokes.”?   Surely the company has tried to show this!

Unfortunately, I don’t know this answer and won’t try to provide some wild speculation (in this particular case at least).  But, I will point out that a possible approach is a well thought out engineering model analysis, hopefully supported by some detailed experiments.  Ok, I have to speculate at least to the point of giving an example.  Suppose that the morphology of the “plaque” buildup or the plaque “deposition” process is altered by the statin.  Then even if the blood level is lower (for LDL), then complications from the disease, say a stroke, is not altered.  

One type of analysis that could be done is to model the deposition process.  You would have to use a PDE and account for the transport of the plaque precursors across the blood flow in arteries, mass transfer to the surface of the blood vessel, an effective reaction rate that causes deposition, perhaps some break off and changes in these caused by alterations in the elasticity of the blood vessel.  Some key “numbers” and even qualitative behavior is probably not known.  So you would want to proceed with caution.

I can cite a very interesting example of this approach, which while it does not appear to have changed the world of cancer therapy, was nonetheless an important paper and perhaps fundamental concept.  Suppose that you develop an anticancer drug treatment that uses a compound that is a strong cancer cell killer with a binding agent that will preferentially bind the cancer killing compound to cancer cells.  Thus you expect that concentrations of your treatment will be much higher in a tumor than in other tissue.  The idea would be that you could increase the effectiveness of a finite amount of the (toxic) compound that is active against cancer cells and reduce the overall patient side-effects.  Now you test an anticancer drug on “tumor prone” rats and find that for low binding there is an average tumor shrinkage of 30%, for medium binding strength the tumor shrinkage is 60% and for strong binding the shrinkage is…. unfortunately only about 15%.  How could this be explained?   In the publication:  C. Graff, K.D. Wittrup, Theoretical analysis of antibody targeting of tumor spheroids: importance of dosage for penetration, and afinity for retention, Cancer Res. 63 (2003) 1288–1296., the authors provide a mechanistically based explanation.  They solved the reaction - diffusion equation for a spherical tumor.  This equation has been used by chemical engineers for decade to analyze catalyst particles.  You can get a Mathematica note book that I developed to try some calculations yourself at ( http://www.nd.edu/~mjm/tumor_calcs_1.nb) .  What you would find as you increase the binding strength, is that the “front” of the diffusion of the drug into the tumor gets sharper.  The sharp front is between the outer tissue of the tumor, which would receive the effect of the drug, and the inner tissue of the tumor, which would be treated.  The result of the analysis is that while there is an advantage to some specific binding, if it is too strong, the active drug will not get to the middle of the tumor.

So what is the take home message?  Statistical analysis is essential to determine if apparent correlations are valid or due to mere random chance.  However, even when correlations are valid, if the mechanism that determines the correlation is not known, then the result either may be useless or at least further generalizations cannot be made.  The tools of engineering analysis can provide a way to investigate possible mechanisms of action.  Just be careful to verify these models at every length and time scale.  Humans and even cells are very complex and hence it is easy to develop a completely meaningless model.

Sunday, January 10, 2010

Random Thought: Engineers do "numbers" better than anyone else

I often tell students and other people that "... engineers do numbers better than anyone ..." including economists, finance people, accountants or even scientists.  The origin of my statement is that we combine knowledge of very strong mathematical tools, that include calculus and differential equations (and hence the ability to construct continuously solvable models -- based on fundamentals or heuristics if necessary) as well as experience dealing with experimental data and hence the ability to interpret information and validate models.  No other academic discipline combines all of this.

Well the validity of this thought was demonstrated this past week.  On Thursday job "creation" data were released for the US and the very unfortunate news was that the economy shed another 85000 non-farm payroll jobs.  This was branded "surprising", as has essentially every economic headline has been for the past year.  Until this time, I had not looked at why they (Economists, I guess) were always "surprised" but I had suspected that they did not understand "noise" (fluctuations) that are always present in data.

Well, this is it!!  Take a look at a bar graph of the job loss data.  Note that in November the losses were essentially zero (actually a month ago this number was a few thousand negative and was revised to + 11,000).  So "economists" had predicted some thousands of positive jobs by simply extrapolating the trend!  If this does not happen then they are "surprised"?(##!!)  If they had asked an engineer, she or he would have said that the trend appears to be decreasing from the really bad months, but if a specific number were needed, it would have come with an uncertainty analysis and some sort of error bars.  (In an engineering situation, any relevant fundamental principles would also have been employed.)   From this point of view, a loss of 85,000 in December would not have produced even a little surprise to an engineer.  Our plants have to operate even if the weather is really hot or really cold, if the feedstocks are not exactly what we want or if we need to make some alterations to increase or decrease production!

So as I was saying, Engineers do numbers better than anyone else!


Friday, January 8, 2010

Random Thought: Post surgical infections,

The New York Times published an interesting article (http://www.nytimes.com/2010/01/07/health/research/07infection.html?ref=science )describing medical studies stating that many of the infections that people get in hospitals are caused by bacteria that they bring in. In these cases, potential solutions are knocking down these bacteria before surgery and this is being done. However, it appeared that more could be done and there was the striking point made that chlorhexidine-alcohol, worked much better for surgical prep than povidone-iodine but the povidone-iodine was used because the cost per patient was $3.50 as opposed to $12 for chlorhexidine - alcohol.


It would be stunning if true: saving $8.50 but raising the risk of infection! It is hard to find an operation that needs a serious surgical prep for which the surgeon's bill would not be more than $5000!


Other than sanity and cost analysis, what is the connection with chemical engineering? A simple model for bacteria growth is that each bacterium splits into two daughters every growth time period. This mechanism leads to exponential growth in the number of bacteria. The time period depends on a lot of things, but 1/2 hour could be used as a nominal value. So why would a better disinfectant be useful? For this growth period, in 24 hours the original number of bacteria will be multiplied by about 163,000. Thus if the original disinfection leaves any significant amount of bacteria, fairly large concentrations could reappear in a day (as a point of comparison: 35 enterococci per 100 ml correlates to an expected rate of 19 diarrheal diseases per 1,000 swimmers -- http://ga.water.usgs.gov/projects/chatm/importance.html). So the need to kill almost all of the bacteria present before surgery is really important!






Random Thought: Football Injuries, bowls, BCS and layoff

Except to the most fervent Alabama fan, the injury to Colt McCoy was a serious blow. I had expected Alabama to be the better team and once the most important player for Texas was knocked out the game, I figured the chances for a good game dropped precipitously. Even though there was some excitement in the second half, this was the case.


As chemical engineer, a couple thoughts came to my mind. The first was did the 30+ day layoff contribute to this particular injury? One of the arguments against a playoff is that the longer season will cause more wear and tear on players and this, along with more exposure, will lead to more injuries (for a small subset of teams). This might be true, but the long layoff between games probably also contributes to injuries -- for quite a few teams. It would be interesting to see some quantitative (statistical) analysis of these questions. If both of these factors contribute, then one could certainly pose an optimization problem for bowl and play-off scheduling that reduces injuries for a given level of revenue (##!!) -- which (of course) must be considered if any changes are to be made.


The injury analysis would be two fold. To get an idea of injuries caused by layoffs, the bowl data would provide a rich source of comparison, say, to a similar temperature profile in the early to middle part of the season. Likewise, wear and tear could be determined, with some weather corrections, by watching the injuries week by week and taking note of byes. In contrast to some of my previous posts, where I note the need for determinism in the analysis, this can be best done simply statistically. Even for the case of a very specific collision involving two or more particular players, slight variations in the turf could mean the difference between a slight strain and complete rupture of a knee ligament.


If such analysis is done, with the follow-on revenue optimization, perhaps we could then decide on logical and scientifically valid grounds if a playoff is a good idea. (If yes, then maybe we could stop using FBS and FCS as sub-designations!)


As a historical note, while I often watch the post Jan 1 bowls, the current system is just not as exciting as the pre-BCS days. In the days when all of the major bowls were on Jan 1 and the AP was going to vote for a champ at the end of the day, all of the bowls mattered. This was because owing to conference alignments, #1 seldom played #2. If your team was ranked 3rd, 4th or maybe even 5th, you were seriously cheering against certain teams either earlier in the day or at the end of the day after your team had won. Often the #1 team did not win and someone else won the title. This was good for Joe Montana's Notre Dame team, but did not work out for our 1993 team -- which I though was one of the all time great College teams. Yes, there were controversies, but it was still more fun than the present!




Tuesday, January 5, 2010

Random Thoughts: Exercise, aging, systems biology, medical systems engineering

The Wall Street Journal published an interesting and informative article today (http://online.wsj.com/article/SB10001424052748704350304574638331243027174.html?mod=WSJ_hps_MIDDLEThirdNews) extolling and explaining (to some extent), the benefits of moderate exercise on health and particularly on reducing the effects of aging. This is, of course, good news to people of my age! As a chemical engineer, there were a couple of interesting issues that generalize beyond this particular article.

First is that an obviously systemic activity (it is hard for just the cells lining your aorta to go for a walk) can be linked to specific molecular consequences, e.g., the lengthening of telomeres -- the strands of DNA at the tips of chromosomes. While I think that there is still some speculation occurring, it is good to see that science can determine the molecular processes that affect important aspects of human health. A second example in the article was reduction in sickness in people who exercise -- presumably by some enhancement of the immune system that occurs due the the elevated blood flow, respiration and physical stresses of exercise. These examples show how both Systems Biology and Molecular biology will be important in the future of human health. To these I see "systems engineering" as a field that can also make significant contributions as I will describe below. (Note that there are researchers working on this strategy right now so I am not claiming originality of these thoughts.)

The second point of interest for me was the complete lack of quantification of effects -- not just that (moderate exercise is walking for some number of minutes, some number of times per week) as compared to more vigorous exercise -- but that results are given statistically as a "X %" reduction in the chances of being stricken with a specific disease. In aggregate, I don't doubt that the statistics are superficially true (well maybe I doubt these a bit), but presenting the results this way makes it appear that the mechanisms and processes involved in both exercise and its effect on disease are stochastic with the outcome a random process with a certain probabilistic value. If I might speculate, I would think that in some cases, there is nothing random at all. A specific deterministic event has to occur for either the disease to take hold or for the prevention mechanism to be effective. Thus certain people are almost certainly going to receive the full value of preventions and others are almost certainly not. Resolving both the correctness of my assertion and producing a useful prediction tool may not occur soon, but is at the crux of what only engineers can bring to health care.

I say regularly, "it is not engineering until we provide numbers!" Well it is also not engineering until the numbers are accurate enough to be useful! Systems biology provides the start of the calculations necessary to provide deterministically-based medical advice, but systems engineering will be necessary to really make a difference to people. The models must be "tuned" to explain medical studies and a synergism developed to suggest the need for additional medical trials.

Let's take the need a bit further. Two older (than me) men who I know well have recently had major, lower abdominal, surgeries for cancers affecting different organs. Both appear to have had all of the cancerous cells removed (along with otherwise useful organs) but unfortunately both contracted post-operative infections. One has fully recovered and I trust the other will as well. However, given the danger of the surgery per-se and the life changing consequences of the operation, a question that would really deserve an answer is if the cancers would have spread and been the ultimate cause of their death at some future date? This is too much to ask right now, but we could again speculate that while there could be some randomness to this process, it would take a very specific deterministic event to occur, for the cancer to spread. Biology, transport phenomena, chemical kinetics and perhaps thermodynamics might be needed to get this ultimate answer -- but a really useful answer it would be!

In addition to the cancer question, these two men were in the X% of surgery patients who contracted post-op infections. It is recognized (and been so since the days of Joseph Lister) that an infection is caused by specific microbes getting into vulnerable tissue. Again, while there is some randomness (say, in the air flow around an open wound or the exact placement and depth of the sutures), this is also largely a deterministic process. Thus in principle the chances of infection for a specific person could be predicted, and, better yet, prevented! Note that the infection rates at medical institutions are different and are publicly available. (e.g., http://www.dhss.delaware.gov/dph/dpc/files/hai_report_2008_final.pdf)

So where does this random thought leave us?

Chemical engineers have very useful, detailed, accurate predictive models of how their chemical processes behave. Million dollar decisions are made daily on buying a tanker of oil that could be needed to "sweeten" a stream at a particular refinery. While now these have considerable fundamental basis, this was developed only through decades of research and comparison with plant performance to verify all of the subcomponents of the computer codes. The earliest models were based on only the mass and energy balances but were still useful for operation and design because of the prudential judgement of the engineers who used them.

We could see the same pathway toward developing useful human process systems models. The fundamental science, molecular, cellular and systems biology is advancing daily. Now is the perfect time for chemical engineers to get moving to make their contributions to this important enterprise!

Monday, January 4, 2010

Random Thought: Declining Body Temp in old(er) adults

This New York Times (http://www.nytimes.com/2009/12/29/health/29real.html) article refers to a recent medical study that found that (apparently) the body temperature of older people declines with increasing age. The question of the relation between (I think oral) and real core temperature aside, the implication of the abstract of the article (http://www.ncbi.nlm.nih.gov/pubmed/18705705 itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_RVDocSum&ordinalpos=3) and and one for an older study (http://www.ncbi.nlm.nih.gov/pubmed/16398904?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_SingleItemSupl.Pubmed_Discovery_RA&linkpos=2&log$=relatedarticles&logdbfrom=pubmed)
is that a "fever" would be occurring at a lower temperature and hence a physician should consider any elevation of temperature as a possible indication of an infection or other inflammatory response.

This is most likely an important observation, but the interesting chemical engineering question that arises is what about the kinetics of the multitude of chemical reactions of life? If we assume Arrhenius kinetics with a nominal activation energy of (what else) 25kcal/mole, we could find that a reaction at 310C would be about 14 slower at 309C and 30% slower at 308C!

The question is if this difference matters to the health of the individual.

If one uses the logic that "evolution" is the ultimate optimization procedure, we would expect that all biological reactions have been optimized in terms of reaction pathways (energetically) and catalytic mechanisms -- with objective function apparently a combination of capital costs -- hardware necessary to run the process -- and operating costs (energy needed to run the reaction). If this is the case, then most processes could be "transport limited" that is the physical processes of diffusion (and convection) of chemical species and heat to and from the specific site of the reaction would limit how fast the reaction occurs. (We have made this argument in our research on transport processes in bone tissue.) If the reactions are transport limited, then the 1 or so degree C temperature change is not likely to matter.

However, most people older than about 30 will tell you that they don't recover from severe physical activity or injuries as fast as when they were younger. So something is changing with age.

Could it be the sustained temperature? Note that with injuries, there is usually significant inflammation which is associated with increased blood flow, so processes may occur at elevated temperatures that do not change with age. However, the question of recovery is interesting (as is the reason muscles hurt after lack of motion) (but I doubt that this is the complete answer: http://www.nytimes.com/2009/12/22/science/22qna.html?ref=science.)

It would be interesting to see if reduced chemical reaction rates because of lower temperature is part of the reason for various declines with age! Further, if there is a way to "warm" people to help them recover!