Sunday, April 02, 2006

Robot for the removal of brain tumours

U0307654 Lian Weiwen, Mervyn

Currently in the medical field, skull surgery to remove brain tumour is a delicate and tedious operation that lasts for an average of 9 hours. It requires 6 hours to drill through the skull and 3 hours to remove the tumour. Moreover, such tumours can be extremely invasive. Thus, their removal is extremely complicated. Furthermore, there may be a need to drill through the temporal bone which is porous and not solid. This increases the complexity of the surgery. Lastly, if the surgeon damages sensitive areas (like the inner ear and facial nerves) of the patient during surgery, it may lead to permanent distortions of the facial features of the patient. Therefore, doctors in this field face a lot of difficulty in performing these operations. (Picture from However, a robot has provided an answer for them! In a joint collaboration between the department of mechanical and production engineering of NTU and the National Neuroscience Institute (NNI), Singapore experts have designed what is believed to be the first robot that is capable of drilling through the temporal bone to remove tumours during surgery. In addition, it is also capable of drilling through the other parts of the skull to remove deep-seated brain tumours. The six-legged computer controlled robot is called Hexapod. With its advent, it is able to shorten operation time to 50 percent. Furthermore, it is also capable of higher precision drills on the skull of the patient. This has translated to a saving in time and an increase in precision. Before the operation, high-resolution magnetic resonance imaging (MRI) scans are done on the patient to obtain information on the location of the sensitive areas of the patient. Such locations vary from patient to patient, so the path that the robot has to take to reach the tumour differs from patient to patient. After this information is obtained, the information is loaded onto a computer to determine the best route to reach the tumour. This route has to avoid the sensitive areas of the patient and has also got to be the shortest route. Next, the robot has to be programmed to take the route that has been planned for it. This program is written on a Linux platform. Besides these, the computer connected to the robot is capable of being preloaded with information obtained from the MRI scans to do a simulation of the actual surgery. This means that there will be greater safety in the procedure because the simulation can be run first and errors can be spotted before the actual surgery. Lastly, the surgeon places physical markers on the patient’s head to guide the robot during the surgery. Currently, trials have been successfully completed on cadavers and trials are being done on animals. In the technology used, there is a high level of accuracy involved. Thus, there is a need to ensure that the control circuitry has to be extremely precise. The controllers have to be well-tuned for movement of the robot and the drilling process. It appears that in this application, the concepts of locomotion and obstacle avoidance have been implemented to enable the robot to avoid the sensitive areas and to move towards its goal, the tumour. In this robot, there seems to be little autonomy. However, this is understandable because in medical applications, if the patient has to pay a high price due to a bug in the software, it would be disastrous. It would have a severe implication on the field of robotics as critics would jump at the opportunity to argue against further research in such a field. Moreover, the feedback from the sensors given to the doctors operating the robot ensures a greater amount of safety in the procedure being performed. Clearly, this robot is a boon to mankind in the field of medical applications. It has helped us to save time and effort. In addition, it has given us greater confidence in executing intricate and delicate medical procedures. Indeed, humans deserve better!

References: 1. 2. 3. 4.


Anonymous said...

The robot does perform better in terms of safety and time to carry out the operation. There is also a need for a doctor to oversee the entire operation. Given this, does it mean that the operation will be much more expensive than one that is carried out by a human doctor? Will an average patient be able to afford for such an operation?

Anonymous said...

U0204840 Lin Ming Zheng

(forgot to put my name in the previous post, sorry for the double posting)

The robot does perform better in terms of safety and time to carry out the operation. There is also a need for a doctor to oversee the entire operation. Given this, does it mean that the operation will be much more expensive than one that is carried out by a human doctor? Will an average patient be able to afford for such an operation?

Security said...


it is very promising to have the robotic to take the role of skilled doctors, and it is a really a great progress in terms of the fact that robotics can do complex intellectual works in replacement of human beings.

But what i concerned here is can we really have the trust in those robotics? i would like to argue that being the most complex part of human body, it is not only the knowledge lacks of the brains and it is also the exact position and time that we have to pay attention to.

What happened if there is a mechanical malfunction when doing the operation? what will happen if the power failure, program failure, the sensor failure... there are lots of issues before we can confidently trust this technique. why not use the kind of precision mechines to other operation in stead of the brain operations.

That's what i want to point out, it may be our desire to have complex tasks accomplished by robotics but we have to consider all the issues that are critical to the liveso of human beings. Maybe we can concentrate on other simple tasks then try out the complex ones.

Industry said...

U0307717 Chew Jian Qiang

Well, I think the overall cost of the operation will be less. I dun think that in an ordinary brain tumour removal surgery, there is only 1 surgeon present. There would have to be multiple specialists present. And the time needed to perform the operation decreases, which would mean the doctors can treat more patients. In terms of economics, supply increases, so price will drop. Just my 2 cents worth. Hehe

Industry said...

u0205081 Chow Synn Nee

I feel very elated after reading this entry. From all the promising results of the simulations and trials on animals, i forsee greater use of such robots for future operations.

As we all know, many operations are tedious and time-consuming and requires precise actions and many a times, the life of the patient really depends a lot on how 'steady' the hands of the surgeon is.

This robot is able to provide the precision required and allows 'route-planning' of how the operation is to be done after obtaining information on sensitive areas of the patient, which is crucial to the whole operation.

Probably with further modifications, such robots would be able to reach hard-to-reach areas to perform operation. For example, tumour at the back of the stomach; this is an area where operation is deemed impossible and often, operation is not advised and patients can only seek radiotherapy or chemotherapy as their form of treatment. With this technology, there is hope that such tumours can be operated on and removed fully. Hence, this robot is definitely a bliss to the medical arena.

Assistive said...

u0205260 - Domingue Jean Michel David

Erm, I still feel uncomfortable about this. Sure you have highlighted a lot about the steps taken to ensure precision and accuracy with the control circuitry and everything. Indeed it'd be tempting to focus on the fact that the robot wont be stressed during the surgery or let 'any' emotions affect him while performing.

At the same time, how much can we rely on a no-fault robot and how much trust can we put in it. Given that the surgery is tedious and long with a human doctor, so we are talking about a trade-off between time (and how if affects performance) with reliability of the robot.

While I sound skeptical about this, I guess that if the developers can perform actual successful operations, that would help to build credibility.

Let's wait for the first actual use of it.

Medical said...

U0307654 Lian Weiwen, Mervyn

To Zhai Ning:
I do agree with you that it is not wise to place implicit trust in robots to do work, especially in medical work where the life of the patient is at stake. Hence, there is a need to perform extensive testing before we use such robots on living humans. This article was published in the computer times in 2003 (mentioned in one of my references). It mentioned that they were conducting animal trials in 2004. Unfortunately, I am unable to find results on the animal trials. With the amount of testing, it appears to me that this technique is safe enough. Moreover, the robot can be used to run on a simulation for verification purposes before the operation. In addition, during the operation, surgeons will be present to guide the robot during the surgery. Hence, there are many layers of precaution available. If a mishap occurs, in my opinion, the authorities would examine if the surgeon performing the operation was negligent and whether he was monitoring the robot. Admittedly, there may be failure of the sensors etc. Under such a case, I would think that the authorities would check if there was extensive testing done on the robot. Having tested on cadavers and living animals, (and hopefully, they did a lot of tests) I would think that these failures would have a low probability or they would not have passed the testing phase.
In the medical field, there are many risks. For example, in developing a new vaccine, it would have to be tested on animals first before it can be administered to live humans. If the probablity of survival is low after taking the vaccine or it has severe side effects, then the medical community will never allow the vaccine to be sold. The key point is that the probability of risk must be small. Risk cannot be eliminated completely. My point is that there would be a small amount of risk in anything we do in the medical field. Howevever, if we are not willing to take any risk, then we would never make any advancement. There would be no polio vaccine or no dialysis machines.
Regarding your question on using other precision machines: This is not quite possible (unless you are referring to the ones that they are currently using) because the tumours that they are referring to are those found at the base of the skull, which is extremely inaccessible. The existing machines that surgeons use currently require them to spend 9 hours on the surgery and they are hoping to reduce this amount of time. Moreover, the skull bone is also one of the hardest bones in the human body. Hence, the difficulty. Moreover, those machines still rely on how steady the surgeon's hands are, so the benefit that this robot offers is greater precision in drilling.

I hope I have clarified any doubts you have. :)

Edutainment said...

u0300657 Yeo Choon Kwang
I do agree that the duration and cost can be reduced when this robot is implemented. I have no doubts about the precision that we will be able to achieve with the current technology.
My concern is the safety aspect of the machine. Simulations and scans are preparations carried out before the actual operation. I understand that the best route is determined and decided before the operation. Path is then pre-programmed into the robot. However, will the doctors be able to stop the robot if some abnormality is discovered during the operation? For example, there may be something the doctor observed that is not captured by the scan during the actual operation. The “best” route determined before the operation is not really the best route. Can the doctor make some last minute changes? If he can make changes to the program in real time and executed the codes from the point where the robot has stopped, will the precision be comprised?

Medical said...

U0204570 Tang Shao Qiang

Using an automated robotic system for surgery is indeed a novel and very exciting medical application. It would benefit complicated surgeries by cutting operative times and offering precise information about the locations of targets. However, certain improvements could be made to the Hexapod. For one, it would be safer if real-time scanning of the hidden tissue structures could be done, so as to ensure, at every instant, that the robotic arm is proceeding in the right direction, adapting to unexpected shifts in patients' position during surgery. In addition, although trials have already been done on cadavers, tests on living beings have yet been done. There are differences between operating on cadavers and live aninmals or human beings. In living beings, considerations have to be given to the perfusion of blood and the monitering of the heart beat and blood pressure. Muscle spasms and the pulsations of blood vessels may make important structures difficult to avoid by the robotic arm that is only guided by pre-planning based on a single initial MRI scan. In the event of a major blood vessel being cut, the Hexapod must be able to respond appropriately and quickly enough to ensure the safety of the patient. The leakage of blood might obscure the robot's view if it uses optical sensing.

Assistive said...

u0204699 Ong Chin Soon

I do believe that the application of robotics in medical field is a great breakthrough in technology. If robots could be engaged to perform complex surgery such as removing a brain tumour with such intricacy and high precision, it indeed opens a whole new application domain for robotics. Nevertheless, like some of the comments made, the implementation of such robots in an actual complicated surgery is definitely something that cannot be taken lightly.

Firstly, robots are afterall robots. They lack the agility, flexibility and ability to react to unforeseen circumstances like human beings. If everything goes on as planned, then the robots would be able to perform the task well or perhaps even better than humans. However, being a complex surgery, the nature of the operation is such that many unexpected situations might just happen in the midst of the surgery, during which the robots would not be able to handle as well as a human surgeon, whom with the many years of medical training, would be able to quickly pinpoint the problem and rectify the situation. It has been mentioned in the blog that the robots would have little autonomy, and that they are largely supervised by human surgeons in the operating theatre. If that is the case, then the extent to which the robots are allowed to perform the operations independently must be planned carefully. On one hand, robots are meant to be as autonomous as possible so as to free up more time for humans to perform other tasks. On the other hand, one refuses to give much autonomy to these robots due to the high risks involve in surgery of such nature. As such, I believe the implementation of Hexapod on a real human surgery should only be done after having weighed the advantage of the robot against other factors, especially cost. If the cost of supervision on the robots is as high as the surgeons performing the surgery themselves, then would the extra high cost of building and introducing the robots justify their implementation in such surgeries?