Scanning, Life Support, Computer Controlled Equipment
Sensors (Analogue and Digital)
Many measurements need to be undertaken on patients in hospital. In order to make this quicker for medical staff the routine recording of data can be done automatically by using sensors. There are two types of sensors:
Data Measured by Sensors and its Uses
Sensors are used in medicine to measure many things such as:
Many measurements need to be undertaken on patients in hospital. In order to make this quicker for medical staff the routine recording of data can be done automatically by using sensors. There are two types of sensors:
- Analogue - Measure analogue quanties that can have an almost infinite set of values such as pressure or temperature
- Digital - Detect digital quantities such as is a switch on,yes or no
Data Measured by Sensors and its Uses
Sensors are used in medicine to measure many things such as:
- Temperature
- Blood
- Pulse
- Blood Sugar
- Brain activity
- Respiratory rate
- Concentration of certain gases in the blood (Blood Gases)
- And many more
Advantages of Using Sensors for Patient Monitoring
Data from sensors can be used to control:
Scanning Devices
Scanning devices are used to model the internal structure of a patient.
MRI
An MRI (Magnetic resonance image) scanner uses magnetic and radio eaves to build a picture of the inside of a patient. MRI scans do not use X-rays meaning that they do no harm to the patient. The patient lies inside a large, powerful cylindrical magnet and powerful radio waves are sent into the patients body. Hydrogen atoms in the patient's body emit radio waves of their own and the scanner picks these up and transforms them into an image. A computer is needed to analyse the data from these radio waves and produce an image on the screen. The image is modelled from the location and the strength of the radio signals the computer receives. The MRI scan is used to produce clear pictures and it is the best technique for doctors to use when they are checking for tumours.
MRI scans can also be used for:
- Measurements are never missed
- Real time monitoring means that nurses can be alerted to any variation can as soon as possible, as opposed to when the next scheduled reading is due to be manually taken
- Gives medical staff more time to focus on other tasks
- Reduces costs
- More accurate readings than manual
- Trends in a patients condition can be spotted
Data from sensors can be used to control:
- Respiration - A ventilator is used to pump air in and out of a patients lungs
- Excretion
- Heart Function - A heart pump can take some work away from the heart if it is stressed or under pressure
- Kidney Function - Dialysis machines are used when kidney functions fail
Scanning Devices
Scanning devices are used to model the internal structure of a patient.
MRI
An MRI (Magnetic resonance image) scanner uses magnetic and radio eaves to build a picture of the inside of a patient. MRI scans do not use X-rays meaning that they do no harm to the patient. The patient lies inside a large, powerful cylindrical magnet and powerful radio waves are sent into the patients body. Hydrogen atoms in the patient's body emit radio waves of their own and the scanner picks these up and transforms them into an image. A computer is needed to analyse the data from these radio waves and produce an image on the screen. The image is modelled from the location and the strength of the radio signals the computer receives. The MRI scan is used to produce clear pictures and it is the best technique for doctors to use when they are checking for tumours.
MRI scans can also be used for:
- Examine the heart and blood vessels for damage
- Examine joints and the spine for damage
- Checking the function of organs such as the spleen
CAT
A CAT (Computerised axial tomography) scanner is a more complex type of X-ray machine. Unlike ordinary X-ray machines, which only send out a single X-ray, CT sans send out several X-ray beams at different angles to the body. The X-rays are then detected after they have passed through the body of the patient, where their strength is measured. The beams that have passed through denser parts of the body will be weaker than those that passed through soft tissue. Therefore the scanner can assemble a picture based on signal strength. Once the signals have been processed the system produces a two-dimensional image on screen. CT scans are also more detailed than normal X-rays and some of the most advanced CT scanners can now produce three-dimensional images. CT scans allow images to be produced that show what a surgeon would only be able to see once opening up a patient. CT scans are usually used for pinpointing tumours in the body and for planning the use radiotherapy for treatment. AS CT scans use more X-rays they can cause side effects and are not performed unless absolutely necessary.
Advantages of Scanning Devices
Disadvantages of Scanning Devices
A CAT (Computerised axial tomography) scanner is a more complex type of X-ray machine. Unlike ordinary X-ray machines, which only send out a single X-ray, CT sans send out several X-ray beams at different angles to the body. The X-rays are then detected after they have passed through the body of the patient, where their strength is measured. The beams that have passed through denser parts of the body will be weaker than those that passed through soft tissue. Therefore the scanner can assemble a picture based on signal strength. Once the signals have been processed the system produces a two-dimensional image on screen. CT scans are also more detailed than normal X-rays and some of the most advanced CT scanners can now produce three-dimensional images. CT scans allow images to be produced that show what a surgeon would only be able to see once opening up a patient. CT scans are usually used for pinpointing tumours in the body and for planning the use radiotherapy for treatment. AS CT scans use more X-rays they can cause side effects and are not performed unless absolutely necessary.
Advantages of Scanning Devices
- Higher cure rate
- Reduces unnecessary operations
- Helps surgeons plan operations
- Faster diagnosis
- Scanning can be done routinely
- Safe - MRI scans
- Can look at internal features without opening a patient
Disadvantages of Scanning Devices
- Sophisticated machinery is computer based and very expensive
- If the machinery breaks it is useless
- Repetitive exposure can be dangerous for patients and staff - CAT
Backup and Recovery Procedures
Scanners use sensors and for scanners to give accurate results the sensors must give the correct readings, thus monitoring the patient correctly. Sensors have a self-test facility in which they check their own readings. This is very important in departments such as the ICU, where an incorrect meaning is literally life or death.Many scans recorded are kept within the patient records so that they can be consulted using any terminal within the hospital. This data is saved in a database that us backed up in real time. Hospitals have their own backup recovery plan that will be put into operation should any of the data or programs be lost or any of the networks or hardware fails to work correctly.
The plan usually uses many aspects of technology including:
New and Future Developments
Modern medicine has started to focus more on patient care outside of hospitals and in particular prevention. For example more routine testing is planned, because if most diseases are detected early enough they can be cured. The development of sensors and communications technology will enable tasks that are normally performed at a GP surgery or hospital outpatients to be performed in people's homes. Advances in sensor technology will create a series of smart sensors which have a small computer chip incorporated into them which will enable data to be not only collected but processed as well. After this step the data will be used from a processor to work an actuator which will allow a measured amount of drugs to be administrated into the patient's body only when the system recognises that it is needed.
Many
Limitations
There are some limitations when using scanning, life support and other computer-controlled equipment which include:
Scanners use sensors and for scanners to give accurate results the sensors must give the correct readings, thus monitoring the patient correctly. Sensors have a self-test facility in which they check their own readings. This is very important in departments such as the ICU, where an incorrect meaning is literally life or death.Many scans recorded are kept within the patient records so that they can be consulted using any terminal within the hospital. This data is saved in a database that us backed up in real time. Hospitals have their own backup recovery plan that will be put into operation should any of the data or programs be lost or any of the networks or hardware fails to work correctly.
The plan usually uses many aspects of technology including:
- Mirrored hard drives
- Backup tapes which use a tape rotation system
- Off-site archiving of data
New and Future Developments
Modern medicine has started to focus more on patient care outside of hospitals and in particular prevention. For example more routine testing is planned, because if most diseases are detected early enough they can be cured. The development of sensors and communications technology will enable tasks that are normally performed at a GP surgery or hospital outpatients to be performed in people's homes. Advances in sensor technology will create a series of smart sensors which have a small computer chip incorporated into them which will enable data to be not only collected but processed as well. After this step the data will be used from a processor to work an actuator which will allow a measured amount of drugs to be administrated into the patient's body only when the system recognises that it is needed.
Many
Limitations
There are some limitations when using scanning, life support and other computer-controlled equipment which include:
- Bandwidth is sometimes limited
- Life support malfunctions can cause the death of seriously ill patients
- Ethical problems such as when to turn off a life support system
- All of this equipment is very expensive and could be better used for prevention rather than cure
Medical Databases
EPR
Electronic patient record keeping
(EPR) had eradicated issues caused by paper record keeping such as:
New and Future Developments of Medical Databases
The NHS National Database
The database currently being developed for the NHS will be the world's larges database containing secure information about a patients healthcare. The new database provides care records which will allow patient healthcare professionals to have quick and reliable access to the records. The new database will contain more than 70 million patients with 400,000 registered users of the system.
Electronic patient record keeping
(EPR) had eradicated issues caused by paper record keeping such as:
- Many staff needed access to the same records simultaneously, thus meaning if there was only one copy of a record this could not be done or would cost time and money to photocopy them
- Storage space was more of an issues as records were physical and needed storage cabinets or drawers
- Some handwritten notes could not be understood by other staff members
- Records frequently went missing
- Notes may have arrived later than required
- Details becoming easier to access in more places
- Improved security
- The data is now easier to back up
- No need to carry and risk loosing records
- Records are available instantly
New and Future Developments of Medical Databases
The NHS National Database
The database currently being developed for the NHS will be the world's larges database containing secure information about a patients healthcare. The new database provides care records which will allow patient healthcare professionals to have quick and reliable access to the records. The new database will contain more than 70 million patients with 400,000 registered users of the system.
The Electronic Prescription Service
The EPS is a new service which will enable prescribers to send prescriptions to patients electronically to a dispenser of the patients choice. This will make the prescription service more convenient and secure for patients and staff. Around 70% of prescriptions issued are repeat medication which can be time consuming due to the current paper-based system however the EPS would allow this process to be significantly faster.
Patient Identification
When a patient arrives at a hospital they are allocated a unique number, which in some cases is their NHS number. This number is important as it provides the key field for the patient database and it is used to distinguish between patients with the same name living at the same address. Patient identification is extremely important in hospitals, so each patient is issued with a wristband with a bar code on it. This system provides the critical first step to making sure that patients are correctly identified. The bar code includes information such as
The EPS is a new service which will enable prescribers to send prescriptions to patients electronically to a dispenser of the patients choice. This will make the prescription service more convenient and secure for patients and staff. Around 70% of prescriptions issued are repeat medication which can be time consuming due to the current paper-based system however the EPS would allow this process to be significantly faster.
Patient Identification
When a patient arrives at a hospital they are allocated a unique number, which in some cases is their NHS number. This number is important as it provides the key field for the patient database and it is used to distinguish between patients with the same name living at the same address. Patient identification is extremely important in hospitals, so each patient is issued with a wristband with a bar code on it. This system provides the critical first step to making sure that patients are correctly identified. The bar code includes information such as
- Name
- D.O.B
- Blood Type
- NHS number, or some other number unique to each patient
Blood Bar Coding and Tracking Systems
Blood transfusions can cause a big risk to patients, if either the blood or the patient is incorrectly identified. Putting the wrong blood into a patient can be fatal. Many infections can also be passed through blood so it is important to identify who gave and received the blood.Blood bags used for the transfusions are routinely bar coded with certain information at the blood bank before being delivered to the hospital.
How Blood Tracking Works
Blood transfusions can cause a big risk to patients, if either the blood or the patient is incorrectly identified. Putting the wrong blood into a patient can be fatal. Many infections can also be passed through blood so it is important to identify who gave and received the blood.Blood bags used for the transfusions are routinely bar coded with certain information at the blood bank before being delivered to the hospital.
How Blood Tracking Works
- Blood taken from the donor is tested to determine its group
- Blood is stored at the National Blood Transfusion Service. Cross matching takes place and bar coded labels are produced during testing and attached to the blood bags.
- Matched blood is sent to the hospital where it is kept in a blood bank/fridge. The details of the blood contained in the bar code are scanned in before the blood bag is placed inside of the fridge.
- Hospital staff remove the required blood by first scanning their ID card. the magnetic locks then unlock on the fridge, allowing the blood to be removed. The identification bar code on the blood is scanned and the blood taken to the ward. This way the blood is tracked in and out of the fridge.
- The patient who requires the blood has their bar code scanned using a handheld PDA. If it matches then the blood transfusion is given.
Advantages of the Blood Tracking System
Use of Internet, Intranets and Extranets
Internet
Hospitals allow staff access to the internet from the organisation's extranet or intranet. Firewalls are used to ensure that people cannot access the hospital system from the internet without permission. The internet can be used by hospitals to:
Extranets
Extranets are networks that use the same technology as the internet but allow people who are not employees of the hospital or trust to access and exchange certain information. They must, however, be authorised to use the extranet which is protected by usernames and passwords.
Extranets can be used by:
Intranets
Intranets are private networks that use the same technology as the internet for then sending of messages of data around the network. only hospital staff are allowed access to the intranet. Many hospitals use intranets as a way of transferring patient data to different devices around the hospital.
- It provides a full audit trail by identifying the member of staff who removed the blood, time/date, location and time spent out of the fridge.
- Eliminates the need for staff to keep paper records - Thus saving time
- Eliminates the need for space to store paper records, which reduces costs
- Cross matching information electronically eliminates the risk of patients being given the wrong blood
- Provides security as only certain members of staff can access the blood in the fridge
- Used for stock control so that the blood to match the patient's needs is always available
- Can find the donor or receiver if there has been a danger of cross-infection
Use of Internet, Intranets and Extranets
Internet
Hospitals allow staff access to the internet from the organisation's extranet or intranet. Firewalls are used to ensure that people cannot access the hospital system from the internet without permission. The internet can be used by hospitals to:
- Email patients the fail to contact by phone
- Communicate with patients using a hospital website
- Allow staff to perform research
- Enable patients to communicate with friends and family
- Send information on lab tests or X-rays requested by doctors in practices or secondary care from hospital or lab servers
Extranets
Extranets are networks that use the same technology as the internet but allow people who are not employees of the hospital or trust to access and exchange certain information. They must, however, be authorised to use the extranet which is protected by usernames and passwords.
Extranets can be used by:
- Suppliers - So they can check that vital drugs and equipment never run out
- Other agencies such as social workers who work with patients outside of the hospital
- GP's to check on the condition of a patient
Intranets
Intranets are private networks that use the same technology as the internet for then sending of messages of data around the network. only hospital staff are allowed access to the intranet. Many hospitals use intranets as a way of transferring patient data to different devices around the hospital.
Distributed Medical Databases
A distributed medical database is a collection of patient information spread over two or more servers in a network. There are many different systems used in health trusts and as a patient you can end up having your details stored on more than one of them. Using internet technology it is possible to build a system that can access all the information about a particular patient even if the information is stored on different servers at different locations. When a healthcare professional needs to see a patient's medical record the user will be unaware the record is in different bits scattered throughout the network.
Distributed databases have the advantage that:
Backup and Recovery Procedures
Medical databases lie at the heart of patient care and it is crucial that this data is kept secure and backed up regularly. Recovery procedures must be in place so that staff can recover any lost data, and it is essential that the procedures are tested to make sure that the system can be recovered in the event of a problem.
Considerations for backup:
A distributed medical database is a collection of patient information spread over two or more servers in a network. There are many different systems used in health trusts and as a patient you can end up having your details stored on more than one of them. Using internet technology it is possible to build a system that can access all the information about a particular patient even if the information is stored on different servers at different locations. When a healthcare professional needs to see a patient's medical record the user will be unaware the record is in different bits scattered throughout the network.
Distributed databases have the advantage that:
- Security can be improved as data is not kept all in one place
- Speed of access is improved because one server does not have to deal with all the requests for information from the user
Backup and Recovery Procedures
Medical databases lie at the heart of patient care and it is crucial that this data is kept secure and backed up regularly. Recovery procedures must be in place so that staff can recover any lost data, and it is essential that the procedures are tested to make sure that the system can be recovered in the event of a problem.
Considerations for backup:
- Whether any downtime is acceptable, if it is not then a RAID system should be used
- How much data needs to be backed up
- Where the data is to be stored, copies should be kept off-site
- How often copies should be taken, this is dependent on how often the data is updated
- When the data should be backed up, usually when the system is being used least such as at night
Expert Systems
Artificial Intelligence
Artificial intelligence is a reasoning process performed by computers, which allows the computer to:
The computer, like a human, is able to learn as it stores more and more data.
Artificial intelligence is a reasoning process performed by computers, which allows the computer to:
- Draw deductions
- Produce new information
- Modify rules or write new rules
The computer, like a human, is able to learn as it stores more and more data.
Neural networks
Neural networks are biological systems that are used by the brain for learning new things. By understanding how the brain works scientists can develop ICT systems making use of artificial neural networks that mimic the way the brain works. The main advantage of artificial neural networks is that they can learn by example just like the human brain, which means they are useful for pattern analysis or data classification.
Ordinary IT systems are good at:
Ordinary IT systems are not good at:
IT systems using neural networks have the following advantages:
However they have the following disadvantages:
How Parallel Processors Work
The main aim of artificial intelligence is to get computers thinking like humans and to do this is necessary to build a computer that works in a similar manner to the human brain. Most computers are used to dealing with one thing at a time, albeit very quickly, which means we are tricked into thinking that it is doing more. The human brain is much more powerful owing to the fact that it consists of more than 1000 billion nerve cells called neurons through which the brain's commands are sent in the form of electronic pulses. This enables the brain to process many pieces of data at the same time, which allows us to think, talk, listen and walk all at the same time. The human brain is able to process data in parallel and this is called parallel processing Specialist computers such as supercomputers are available that carry out parallel processing and these computers are used for very complex computer tasks where many millions of individual items of data need processing very quickly, such as in the production of weather forecasts. Using a single processor the problem is programmed as a series of instructions which are processed by the processor in turn.
Neural networks are biological systems that are used by the brain for learning new things. By understanding how the brain works scientists can develop ICT systems making use of artificial neural networks that mimic the way the brain works. The main advantage of artificial neural networks is that they can learn by example just like the human brain, which means they are useful for pattern analysis or data classification.
Ordinary IT systems are good at:
- Fast processing of data
- Obeying a set of instructions given as the program code
Ordinary IT systems are not good at:
- Adapting to circumstances
- Dealing with data in parallel (Computers like to process in a linear fashion)
- Dealing with data that is imprecise or contains errors
IT systems using neural networks have the following advantages:
- Good where algorithms cannot be developed
- Where there are plenty of examples a system can learn from
- Where a structure can be identified from existing data
However they have the following disadvantages:
- Only suited to certain tasks
- The examples used to teach must be chosen carefully, otherwise time is wasted and the system can produce unpredictable results
- Because the neural network learns on its own its operation can be unpredictable
How Parallel Processors Work
The main aim of artificial intelligence is to get computers thinking like humans and to do this is necessary to build a computer that works in a similar manner to the human brain. Most computers are used to dealing with one thing at a time, albeit very quickly, which means we are tricked into thinking that it is doing more. The human brain is much more powerful owing to the fact that it consists of more than 1000 billion nerve cells called neurons through which the brain's commands are sent in the form of electronic pulses. This enables the brain to process many pieces of data at the same time, which allows us to think, talk, listen and walk all at the same time. The human brain is able to process data in parallel and this is called parallel processing Specialist computers such as supercomputers are available that carry out parallel processing and these computers are used for very complex computer tasks where many millions of individual items of data need processing very quickly, such as in the production of weather forecasts. Using a single processor the problem is programmed as a series of instructions which are processed by the processor in turn.
Expert Systems
What is an Expert System?
An expert system is an ICT system that uses artificial intelligence to make decisions based on data supplied in the form of answers to a question. This means that the system is able to respond in the way that a human expert in the field would come to a conclusion. A good expert system is one that cab match the performance of a human expert in the field.
The Three Components of an Expert System
What is an Expert System?
An expert system is an ICT system that uses artificial intelligence to make decisions based on data supplied in the form of answers to a question. This means that the system is able to respond in the way that a human expert in the field would come to a conclusion. A good expert system is one that cab match the performance of a human expert in the field.
The Three Components of an Expert System
- Knowledge Base - A huge organised set of knowledge about a particular subject. It contains facts and also judgmental knowledge, which gives it the ability to make a good guess, like a human expert.
- Inference Engine - A set of rules on which to base decisions and most of these rules have the "if-then" structure. It is the part of the expert system that does the reasoning by manipulating and using the knowledge in the knowledge base.
- A User Interface (UI) - The UI presents questions and information to the operator and also receives answers from the operator.
How Expert Systems Work
Expert systems use a problem-solving model that organises and controls the steps that need to be taken to solve a problem. One part of this model is the use of the IF-THEN rule to form a line of reasoning. These problem solving methods are provided as part of the inference engine. The knowledge base is all the information an expert uses when arriving at a decision of diagnosis. The more knowledge and expert use a human expert has in the field, then the more knowledge can be added to the knowledge base. Because knowledge is often incomplete or uncertain, a rule can sometimes have a confidence factor or weight. A user of the expert system is asked a series of questions which may include the results of a test and on the basis of the answers, the expert system can arrive at an answer.
Expert System Shells
There are two ways to create an expert system. You can:
An expert system shell is a generic piece of software used for creating expert systems. The word generic means that the user can use the software in different was for different applications. The system shell consists of an inference engine and the user interface but the knowledge base is not supplied. Instead the knowledge base is created by the person/expert who is building the system.
The reason the word "shell" is used is because the user has to build the knowledge base themselves, so the expert system is not considered complete until this is done. The shell enables people who know little about programming to create expert systems, although you still need to know quite alot about experts to create them anyway. This is why many expert systems are created by knowledge engineers.
Software Languages
Programming languages have been developed that are specialist languages for the development of expert systems. These languages use English-like statements which are facts along with results and questions.
Two software/programming languages used for the development of expert systems are:
These languages can also be used to build systems that make use of artificial intelligence.
Medical Uses of Expert Systems
Much of medicine is based around the making of decisions based around certain facts so expert systems find many uses in this field. Here are details of some of the expert systems used for patient healthcare.
MYCIN
MYCIN was the earliest expert system used in medicine and uses a patient's blood tests and results to enable doctors to pinpoint the correct organism, from many, that is responsible for a blood infection. Until the development of this expert system, it was necessary to grow a culture of the infecting organisms, which took 48 hours, and if doctors waiting until the results came through, the patient could be dead. By identifying the correct organism using the system, specific drugs can be given to the patient that will treat that particular organism. Using the system means that junior doctors can make an accurate diagnosis with the accuracy of an expert doctor in the field and this means that patients can be out on the correct medication immediately.
Features of the MYCIN system include:
The success of any expert system can be judged by comparing the system to a human expert. With MYCIN they found that the system compared equally with a human specialist in blood infections and better than a GP.
NEOMYCIN
Neomycin is the name of another expert system used in medicine which was developed in order to train doctors. The system took the doctors through various example cases, checking doctors' conclusions and the explaining, if the went wrong, where they went wrong.
Advantages:
Disadvantages:
Expert systems use a problem-solving model that organises and controls the steps that need to be taken to solve a problem. One part of this model is the use of the IF-THEN rule to form a line of reasoning. These problem solving methods are provided as part of the inference engine. The knowledge base is all the information an expert uses when arriving at a decision of diagnosis. The more knowledge and expert use a human expert has in the field, then the more knowledge can be added to the knowledge base. Because knowledge is often incomplete or uncertain, a rule can sometimes have a confidence factor or weight. A user of the expert system is asked a series of questions which may include the results of a test and on the basis of the answers, the expert system can arrive at an answer.
Expert System Shells
There are two ways to create an expert system. You can:
- Build the expert system from scratch using a software language suited to this task
- Use a piece of development software called an expert system shell
An expert system shell is a generic piece of software used for creating expert systems. The word generic means that the user can use the software in different was for different applications. The system shell consists of an inference engine and the user interface but the knowledge base is not supplied. Instead the knowledge base is created by the person/expert who is building the system.
The reason the word "shell" is used is because the user has to build the knowledge base themselves, so the expert system is not considered complete until this is done. The shell enables people who know little about programming to create expert systems, although you still need to know quite alot about experts to create them anyway. This is why many expert systems are created by knowledge engineers.
Software Languages
Programming languages have been developed that are specialist languages for the development of expert systems. These languages use English-like statements which are facts along with results and questions.
Two software/programming languages used for the development of expert systems are:
- PROLOG
- ASPRIN
These languages can also be used to build systems that make use of artificial intelligence.
Medical Uses of Expert Systems
Much of medicine is based around the making of decisions based around certain facts so expert systems find many uses in this field. Here are details of some of the expert systems used for patient healthcare.
MYCIN
MYCIN was the earliest expert system used in medicine and uses a patient's blood tests and results to enable doctors to pinpoint the correct organism, from many, that is responsible for a blood infection. Until the development of this expert system, it was necessary to grow a culture of the infecting organisms, which took 48 hours, and if doctors waiting until the results came through, the patient could be dead. By identifying the correct organism using the system, specific drugs can be given to the patient that will treat that particular organism. Using the system means that junior doctors can make an accurate diagnosis with the accuracy of an expert doctor in the field and this means that patients can be out on the correct medication immediately.
Features of the MYCIN system include:
- Uses around 500 rules
- Will ask for further information such as "has the patient recently suffered burns?"
- Will ask for certain lab tests if needed
- Will suggest the best medication to treat the infection
- Will give a list of drugs with probabilities of the drug working
- Will make a final choice of the drug based on a series of questions the system asks about the patient allergies.
The success of any expert system can be judged by comparing the system to a human expert. With MYCIN they found that the system compared equally with a human specialist in blood infections and better than a GP.
NEOMYCIN
Neomycin is the name of another expert system used in medicine which was developed in order to train doctors. The system took the doctors through various example cases, checking doctors' conclusions and the explaining, if the went wrong, where they went wrong.
Advantages:
- Consistency - Provide a consistent answer for repetitive decisions
- Cheaper - Cheaper than using a human expert such as a doctor or consultant
- The system can consult a larger pool of knowledge compared to a human
- The system is available 24/7 and is never on holiday or off sick when needed
- The computer uses all the information it has, unlike a human expert who may forget and make mistakes
Disadvantages:
- No Common Sense - If the MYCIN system was used with a patient who had been shot and was bleeding to death, the system would only look to stop infection
- Can Make Absurd Errors - If the data is incorrectly input then the drugs prescribed can be absurd.
- Not Able to Provide a Response - Human experts can produce creative responses in certain situations, which an expert system is unable to do
- Not able to realise when no answer is available to a problem
- Relies on the rule and knowledge base being correct. Any mistakes in these could cause incorrect diagnoses