jaredy

Ayun. inupdate ko ang schema natin.
Hindi naman talaga kailangan yung deleteAll or whatever dahil may on delete cascade naman…

ayan ginawa ko na yung sql para jan.

Hindi rin supported ng gwt ang arraylist… tsaka may mga sql syntax pala dun sa binagong daoClasses kaya sya nagexceptions…

ayun lang.

==============================

1) http://www.chiroweb.com/archives/15/10/09.html

“There are very few pathognomonic signs and symptoms that illicit the immediate neurotoxic diagnosis. Variations in symptom complex may mimic disorders of psychiatric, metabolic, nutritional, inflammatory, or degenerative natures. Common complaints may include numbness, tingling, pain, weakness, dizziness, and forgetfulness.”

2) http://www.inchem.org/documents/ehc/ehc/ehc72.htm

International Programme on Chemical Safety. “Principles of Studies on Diseases of Suspected Chemical Etiology and their Prevention”

3) http://books.google.com.ph/books?id=BfdighlyGiwC&pg=PA111&lpg=PA111&dq=pathognomonic+signs+and+symptoms+in+toxicology&source=web&ots=Kr85wLv7Wb&sig=Ah8RiKRb7ky2Zg4inJ2T4pGEtQ4&hl=en

Richard C. Dart. “Medical Toxicology.”

Expert System for Poisoning

Data Dictionary

SQL files ::

sinend ko na lang sa groups yung ibang tapos na…to follow na lang yung iba. ^_^

0530pm, sa isang magandang room sa DILC, 2nd floor, nag-nosebleed na naman kami sa mga medical terms and other topics na kelangan i-research…hehehe.

here’s the minutes

word for the day :: PATHOGNOMONY

wahaha….pathognomonic ang mas common na term…

Dahil parang wala tayong mapapala doon sa uber-simpleng algo natin for diagnosis, pinag-usapan namin kanina ni Dy kung gamitin na natin yung decision trees sa algo natin. Bali yung algo ay something like this:

1) Upon inclusion of new toxicological class, add the toxicological class details to the toxicological class table that lists its clinical manifestations
2) Convert the toxicological class table to a decision tree
3) If necessary, divide the decision tree into subtrees
4) Convert each subtree into a CLIP file
5) Clear and reload all CLIP files for the ESP Inference Engine

So kailangan nating i-formalize at i-substantiate ang algong ito, particularly nos. 2 - 4, if ever

SETH: an expert system for the management on acute drug poisoning

Jean-Michel DROY, MD (1) , Stéfan J. DARMONI, MD, PhD (2) , Philippe MASSARI, MD (1), Thierry BLANC, MD, M. Sc, (3) , Fabienne Moritz, MD, M. Sc, (1), Jacques LEROY, MD (1)

ABSTRACT

The aim of SETH is to give specific advice concerning the treatment and monitoring of drug poisoning.

… which is almost exactly the end-goal of our system except that we are more focused in the recognition of the toxicant in each poisoning case. Including treatment and monitoring advice is secondary and, I think, even optional.

Currently, the data base contains the 1153 most toxic or most frequently ingested French drugs from 78 different toxicological classes.

… wow… how many toxicological classes do we plan to support? I suggest that we limit ourselves to around 20 of the more common toxicological classes encountered here in the Philippines. We can use the “black book” from the National Poison Control Center as our reference. But then, we would really have to design our system as easily updatable by a knowledge engineer…

The SETH expert system simulates expert reasoning, taking into account for each toxicological class, delay, clinical symptoms and ingested dose.

… I think that we would really have to decide if we are going to take “ingested dose” into account when making the diagnosis. I mean, do patients usually know how much poison they have taken in? Do the ones who brought him to the hospital usually know that piece of vital information? I do not know. If so, we would have to accommodate two scenarios in our system: known ingested dose and unknown ingested dose (just like the cases here in SETH). I suggest that we just stick to the unknown-ingested-dose case.

It generates accurate monitoring and treatment advice, addressing also drug interactions and drug exceptions.

… drug interactions and drug exceptions? Hmmm…. I guess that would be difficult for us to take into account, especially because we do not have a definite medical expert to collaborate with.

SETH is then daily used by residents as telephone response support on drug poisoning.

… this is quite close to how I envisioned our system would be used.

INTRODUCTION

Reasoning systems can go further: (a) in a known intoxication to give non-toxicologist physicians better advice according to drugs ingested, clinical manifestations and delay, (b) in an unknown intoxication to identify products according to clinical manifestations.

… this is what I was writing about earlier. What do you think? Two scenarios or just one?

MATERIAL AND METHODS: Decision in toxicology

In most cases of acute drug poisoning (85% of the overall intoxication), drugs and ingested quantities are hypothetically known from patient or family report.

… is this also the case here in the Philippines? Because if the toxicant is already known, our system would not be that beneficial, don’t you think? The attending physician can immediately proceed with treatment based on the presumed toxicant.

In case of poisoning with a single drug, toxicological reasoning is simple: the intoxication can be confirmed by the presence of one or more clinical findings. It can be eliminated if one very sensitive sign is absent, such as bradycardia in beta-blocker poisoning, after checking that the patient has been examined within a time frame compatible with the drug’s toxicokinetics.

… now, how do we know the “sensitive signs” or toxidromes? We really need to collaborate with a toxicologist… how can we arrange that? If only we pursued the toxicologist from the National Poison Control Center… A drug’s toxicokinetics is also a problem… do we take “time” into account when assessing a particular case? That would assume that we know the time of toxicant ingestion of the patient from himself/herself or the ones who brought him in. I suggest that we take “time” into account because some of the toxicants from the “black book” have that piece of information. I just don’t know how it would factor in in our inferencing process.

Poisoning severity depends on the type of drug, the quantity ingested which can be assessed by questioning, the clinical context of the patient, his/her medical history, and the possible ingestion of other toxic, such as alcohol.

… I guess we also haven’t considered if we are going to consider both acute and chronic cases of poisoning. Would the diagnostic process for the two cases differ? Now, out of all this criteria, the only certain criterion that we would consider is the clinical context of the patient. Regarding the possible ingestion of other toxicants, that has got me into thinking that maybe we should also take the assumed typed of drug and the ingested dose in some cases… what do you think? Should we also consider the known-ingested-dose case for our system? Hmmm…

In case of poisoning with more than one drug or with a drug containing more than one product (the most frequent drug poisoning in adult), toxicological reasoning is far more complex. Textbooks (4-6), product and case data bases and the advice of a Poison Control Center (PCC) are often necessary to handle such a patient.

… would we have access to the necessary information for handling this case? I deem it too difficult… Maybe we should just assume a single toxicant for each poisoning case. We are, after all, giving a list of possible toxicants in the order of degree of possibility as output. Maybe a long list of almost equally possible toxicants denote a case of multiple toxicants.

In case of an intoxication with unknown drugs, the only data available to the physician are signs and context. From these data, he/she must deduce the list of toxicological classes which can explain all the signs and the context.

… we can limit our system to just this… We can always assume that the any information about the toxicant is unknown. Attending medical personnel can then just use our system for gathering monitoring and treatment information when the toxicant is already known, possibly skipping the diagnostic process altogether.

[TO BE CONTINUED...]

ESP (Expert system for poisoning) :: A Clinical Decision Support System for the Diagnosis and Management of Poisoning

*Knowledge Base at Inference Engine pa lang ito

visit our new webpage here.

try lang yan so wala lang.

October 5 - finalize project name with advisers

October 11 - 12 noon
short paper draft - acm format (hard and soft copies with latex source - email to advisers)

October 15 -  12 noon
web pages

peer evaluation form (individual, fill up in secret, submit in a sealed envelope signed
across flap)

short paper final - acm format (hard and soft copies with latex source - email to advisers)