Physics Physics in medical physics? Leaving medical physics?

[Quantum_Cthulhu]

I'm a medical physicist in training from the UK. One year away from being fully qualified. I feel like a technician. I have literal anxiety as I watch my maths and physics skills go down the drain (I try in my spare time, but my spare time is not much).

Has anyone (training or working to be a medical physicist) managed to collaborate with a university and get some nitty gritty maths/physics done? Some mathematical modelling? Solving a differential equation? And I'm not talking about "research" that involves developing more QA or (god forbid) more phantoms to scan.

If not, has anyone successfully transitioned out of medical physics? If so where did you end up? I feel the skillset is too niche it would be hard.

 

[Choppy]

Sure. I'm a medical physicist and do a fair amount of research. I can't really speak to the UK. In Canada it's quite common for medical physicists to have adjunct university appointments and be involved in some level of research and teaching. It certainly can be more challenging if you're at a smaller clinic where you don't have access to many of the newer technologies though, and if you have to lead the project yourself. But there are usually opportunities if you seek them out. Normally when I go to conferences, I come back full of different ideas for projects and then inevitably get upset when I realize that I can't possibly work on all of them.

 

[gleem]

I'm a medical physicist in training from the UK. One year away from being fully qualified. I feel like a technician. I have literal anxiety as I watch my maths and physics skills go down the drain (I try in my spare time, but my spare time is not much).

True there is a lot of routine in medical physics. I will be speaking from a nonuniversity affiliated clinical perspective in the US. I am retired so my comments may not precisely reflect the current situation. In my setting, I spent routinely 10 hrs in the clinic each day. This allowed for time to attend meetings and read journals. The situation that you put yourself in is a combination of the expectations of the administration and how you can achieve satisfaction. In the US MPs are paid well so the expectations of the administration are high and they do not typically pay one to do research although they do like to see their name on publications.

In the US only about 10% of physicists do research and mostly at/with university-affiliated departments. For theoretical projects your most precious requirement is time. As I said above one can find some time for one own endeavor, but will probably include some personal time. The question is how bad do you want it? If you can cobble together a few hours on your employer's time but are not willing to do the same on your time then it may not work. In the US besides the annual meeting of the American Association of Physicists in Medicine, we have affiliated regional medical physics organizations that meet regularly in which you can establish collegial relationships with other MPs. These meetings are usually located so most members can attend by automobile.

So what do your UK colleagues say?

If you feel you need to change your goals do it as soon as possible.

And I'm not talking about "research" that involves developing more QA or (god forbid) more phantoms to scan.

I'm sorry you feel that way such as these are so important to consistent quality procedures.

 

[Quantum_Cthulhu]

Sure. I'm a medical physicist and do a fair amount of research. I can't really speak to the UK. In Canada it's quite common for medical physicists to have adjunct university appointments and be involved in some level of research and teaching. It certainly can be more challenging if you're at a smaller clinic where you don't have access to many of the newer technologies though, and if you have to lead the project yourself. But there are usually opportunities if you seek them out. Normally when I go to conferences, I come back full of different ideas for projects and then inevitably get upset when I realize that I can't possibly work on all of them.

Thanks for your reply Choppy. Does the research you're involved in require mathematical modelling/calculus/ODEs/PDEs? I'm just looking to use techniques I enjoyed in undergrad. Would love any papers to look at. Even if it's just closely related research for anonymity purposes.

 

[Quantum_Cthulhu]

True there is a lot of routine in medical physics. I will be speaking from a nonuniversity affiliated clinical perspective in the US. I am retired so my comments may not precisely reflect the current situation. In my setting, I spent routinely 10 hrs in the clinic each day. This allowed for time to attend meetings and read journals. The situation that you put yourself in is a combination of the expectations of the administration and how you can achieve satisfaction. In the US MPs are paid well so the expectations of the administration are high and they do not typically pay one to do research although they do like to see their name on publications.

In the US only about 10% of physicists do research and mostly at/with university-affiliated departments. For theoretical projects your most precious requirement is time. As I said above one can find some time for one own endeavor, but will probably include some personal time. The question is how bad do you want it? If you can cobble together a few hours on your employer's time but are not willing to do the same on your time then it may not work. In the US besides the annual meeting of the American Association of Physicists in Medicine, we have affiliated regional medical physics organizations that meet regularly in which you can establish collegial relationships with other MPs. These meetings are usually located so most members can attend by automobile.

So what do your UK colleagues say?

If you feel you need to change your goals do it as soon as possible.


I'm sorry you feel that way such as these are so important to consistent quality procedures.

In the UK we are not even paid well, but neither are most jobs. Your $250k salaries are equivalent to our ~£50k. By the time you rebalance for the cost of living, it still doesn't pay off. In the UK, as a trainee, as well as integrating into the clinical workload, we are the ones left to take on all the research projects. In fact, I have to produce a big, year-long "innovation" project. The issue is we are either:

1.) "Researching" QA. Can we buy a new phantom? Can we make some code to automate this other QA?
2.) How well are we performing (audits)? Adaptive radiotherapy for example - let's pull some numbers out and see if we are doing better than we were or lets see if we can reduce PTV margins.
3.) "Assisting" medical doctors and PhD students in biomedical/neuroscience in their projects. By assisting I mean, with MRI for example, if there is a cool research project - the MRI physicist's job is to make sure it's safe, but at least their names on the paper I guess.

So after that realisation, I looked to research papers. I am specialising in MRI and radiotherapy. In MRI the big thing right now is AI and deep learning (not physics but at least it involves some maths and programming). I look at the research papers - all biomedical engineers (Medical physics doesn't give me the skillset to get involved and yes I have done courses in deep learning in my spare time to try). So I turn to radiotherapy - physicists outside of medical physics study and develop FLASH radiotherapy. So I go to conferences and I am met with 1.), 2.) and 3.). IPEM is to us what AAPM is to the US. I have quite literally been in a conference where MRI physicists argued why they even bother doing some redundant QC.

I have given my spare time to create code such as discreetly solving the LBTE vs stochastically via monte carlo to show the difference in treatment planning algorithms. It wasn't worth the extra stress of no free time. The student's don't need to know the physics that in-depth. So it was not used in the curriculum.

Yes, these things we do are all very important to a hospital. Is it physics? No. Does it involve physics? I hear the argument that understanding the fundamentals is what makes it physics - it's important when things go wrong. One could argue in the same vein that understanding the physics of washing dishes is beneficial as you understand to clean the dishes you need to create friction. Lastly, when things go wrong, it's most likely a call to the vendor these days. Not back in my manager's day so he reminds me.

 

[gleem]

In the UK we are not even paid well, but neither are most jobs. Your $250k salaries are equivalent to our ~£50k. By the time you rebalance for the cost of living, it still doesn't pay off. In the UK, as a trainee, as well as integrating into the clinical workload, we are the ones left to take on all the research projects. In fact, I have to produce a big, year-long "innovation" project. The issue is we are either:

I'm confused. Is this comment with regard to your training program? Did you not know all these issues before you entered this program? According to https://www.erieri.com/salary/job/medical-physicist/united-kingdom the entry-level salary for an MP is £86K ($(US)106K) for a PhD. What is your degree?

One could argue in the same vein that understanding the physics of washing dishes is beneficial as you understand to clean the dishes you need to create friction.

Really? About ten years ago we had a similar discussion with one who criticized the term medical physics. I think @Choppy remembers.

I was wondering, are your aspirations in theoretical physics? Most of the MPs I knew did their degree work in experimental physics. Physicist are recruited into MP because of their skills first and knowledge of physics second. Those skills include problem-solving, analytical skills, attention to detail, and understanding of physical principles and their technology application. Physicists work with physicians to acquire, site and use equipment with suitable characteristics for the task, Physicians depend on the MP to make good recommendations and to ensure that they are the best possible for the circumstances and continue to work that way. Something that entry MP may not know is that they work on a day-to-day basis with many different healthcare professionals including physicians, nurses, technologists, administrators, and even janitors (sometimes). They must have good communication skills and a great amount of flexibility. It is not a job for everybody with a PhD in physics.

 

[Quantum_Cthulhu]

I'm confused. Is this comment with regard to your training program? Did you not know all these issues before you entered this program? According to https://www.erieri.com/salary/job/medical-physicist/united-kingdom the entry-level salary for an MP is £86K ($(US)106K) for a PhD. What is your degree?

So within the UK a PhD is irrelevant. Maybe an employer would see it as a slight advantage? Majority of the time not. In fact, talking to peers my age and slightly older, the tendency is to view a PhD as a waste of time as 1.) It does not guarantee anything, job or promotion. Majority of all hires (from training to consultant level) have all had no PhD. So the time and money lost is very discouraging (despite the fact I want to do a PhD).

When you enter one of the 3(?) training schemes within the UK that is required to become a medical physicst (similar to residency and ABR exams I think in the US?). To become a medical physicst you will start at NHS band 6 (£37k to £39.4k after 2 years). This involves an MSc as well as in hospital training.

Then you will likely finish your trainkng in 3-3.5 years and become a fully qualified registered medical physicst on band 7 (£46k-£52k after 5 years).

If you have been keeping on track over the 5 years since you become fully qualified, you will have filled out an extensive portfolio to submit in order to become a medical physics expert (MPE). This allows you to apply (if positions are available) to band 8a jobs (£53 - £60.5k after 5 years). On average, it takes about 7-8 years after being fully registered.

Now if you do more portfolios and gain extra titles you may then apply after several more years to a band 8b job which is very hard to come by (£62k - £72k after 5 years). Band 8b jobs are those that lead sub departments. For example, within radiotherapy, you could be the "head of treatment planning," or "head of brachytherapy physics." Since these are so hard to come by, the majority will cap out at 8a for the rest of their career.

Now, if someone retires or dies and you managed to get to 8b... Then if you're lucky you'll get an 8c job (£74k - £85.6k after 5 years). This job means you are the head of radiotherapy, or nuclear medicine etc. so there will be between 2-4 of these roles within a whole board or trust.

Lastly, there is an even rarer job - being the head/director of all of medical physics for your NHS board/trust. This is a band 8d job (£88k - £101k). There is sometimes not even 1 of these per board/trust. There may be less then 15 in the whole of the UK. Basically the majority stay at 8a, get an 8b if lucky. Past 8a you are essentially a manager.

Bottom line, an entry level PhD (assuming they have done all the training etc.) will start at band 7. I am training with several people who have PhD's. The only difference is I'm 4-8 years younger and we earn the same amount (band 6).

Really? About ten years ago we had a similar discussion with one who criticized the term medical physics. I think @Choppy remembers.

I was wondering, are your aspirations in theoretical physics? Most of the MPs I knew did their degree work in experimental physics. Physicist are recruited into MP because of their skills first and knowledge of physics second. Those skills include problem-solving, analytical skills, attention to detail, and understanding of physical principles and their technology application. Physicists work with physicians to acquire, site and use equipment with suitable characteristics for the task, Physicians depend on the MP to make good recommendations and to ensure that they are the best possible for the circumstances and continue to work that way. Something that entry MP may not know is that they work on a day-to-day basis with many different healthcare professionals including physicians, nurses, technologists, administrators, and even janitors (sometimes). They must have good communication skills and a great amount of flexibility. It is not a job for everybody with a PhD in physics.

It is my fault for not doing more thorough research, but essentially I just thought as research is a part of the job description that we are given, and we are titled physicsts, our research would involve more physics. So in that regard I thought it would be more theoretical.

I essentially seem this as a good trade off to:

1.) Actually make an immediate and impactful different in people lives compared to academia physics (check).
2.) Have better job security and pay than academia (check?)
3. Still get to conduct a little physics/mathematically involved research, although not my main priority (not checked).

I suppose my frustration comes from the title itself. I just want to be expected to do some physics. Even if it's directly relevant. Like some particle physics in relation to radiotherapy. Some EM related to MRI, nuclear physics related to nuclear medicine etc.

The points you have made about the translational skills are great and I agree. I just wish the job was titled differently so I didn't expect there to be a least minimal physics to be done, even if only 5% of the job.

 

[Choppy]

I think there's a lot to unpack here. I'm not going try to convince you that medical physics is necessarily a math-heavy field, or that you're going to be solving the Schrodinger equation on a regular basis. It's an applied physics field where the day-to-day issues tend to fall more along the lines of... output on one of your units changed by x% last night, should you take the machine out of service?... or help me plan a bilateral breast treatment with a tissue expander in the way... or validate that the fancy new AI contouring algorithm is doing what it's supposed to be doing... or run the shielding calculations for the new vault you're centre is constructing (mostly spreadsheet-based attenuation calculations), etc., etc. Some days it feels like you're simply the universal translator so that physicians, IT, radiation therapists, electronics engineers, vendors and finance can all talk to each other.

Thanks for your reply Choppy. Does the research you're involved in require mathematical modelling/calculus/ODEs/PDEs? I'm just looking to use techniques I enjoyed in undergrad. Would love any papers to look at. Even if it's just closely related research for anonymity purposes.

It has, sure. Though to be fair I've come through the Canadian system where there has always been a lot more emphasis on research in medical physics. Most medical physicists here tend to work in larger centres with academic roles and responsibilities. For example, I had the opportunity early in my career to work on a team that developed one of the first working linac-MRI machines, and there was certainly a lot of E&M-heavy work involved in that.

I hear the argument that understanding the fundamentals is what makes it physics - it's important when things go wrong. One could argue in the same vein that understanding the physics of washing dishes is beneficial as you understand to clean the dishes you need to create friction.

Except that when you wash dishes wrong it doesn't wallop a cohort of patients with a toxicities that will degrade their quality of life.

I have given my spare time to create code such as discreetly solving the LBTE vs stochastically via monte carlo to show the difference in treatment planning algorithms. It wasn't worth the extra stress of no free time. The student's don't need to know the physics that in-depth. So it was not used in the curriculum.

You're working in MRI and radiotherapy and you've created your own linear Boltzmann solver? You might be interested in work like this, where they've been working to develop a Boltzmann solver that incorporates magnetic fields...
https://aapm.onlinelibrary.wiley.com/doi/abs/10.1118/1.4937933
https://aapm.onlinelibrary.wiley.com/doi/full/10.1118/1.4905041
https://iopscience.iop.org/article/10.1088/1361-6560/aa959a/meta
https://iopscience.iop.org/article/10.1088/2057-1976/ab6e15/meta
Here's one where they're using a neural network to predict the influence of a magnetic field on the dose distribution...
https://www.sciencedirect.com/science/article/abs/pii/S1120179720302763
Or here's a single author work that incorporates protons...
https://iopscience.iop.org/article/10.1088/1361-6560/acf4de?utm_source=wiley&getft_integrator=wiley

 

[Quantum_Cthulhu]

It has, sure. Though to be fair I've come through the Canadian system where there has always been a lot more emphasis on research in medical physics. Most medical physicists here tend to work in larger centres with academic roles and responsibilities. For example, I had the opportunity early in my career to work on a team that developed one of the first working linac-MRI machines, and there was certainly a lot of E&M-heavy work involved in that.

Apologies for the late reply. Firstly, this is very impressive and interesting.

You're working in MRI and radiotherapy and you've created your own linear Boltzmann solver? You might be interested in work like this, where they've been working to develop a Boltzmann solver that incorporates magnetic fields...
https://aapm.onlinelibrary.wiley.com/doi/abs/10.1118/1.4937933
https://aapm.onlinelibrary.wiley.com/doi/full/10.1118/1.4905041
https://iopscience.iop.org/article/10.1088/1361-6560/aa959a/meta
https://iopscience.iop.org/article/10.1088/2057-1976/ab6e15/meta
Here's one where they're using a neural network to predict the influence of a magnetic field on the dose distribution...
https://www.sciencedirect.com/science/article/abs/pii/S1120179720302763
Or here's a single author work that incorporates protons...
https://iopscience.iop.org/article/10.1088/1361-6560/acf4de?utm_source=wiley&getft_integrator=wiley

Secondly, this ^ is exactly what I dreamed I would be able to do at least 5% of the time as a medical physicist. I feel like if I could get involved in this type of thing, I could truly live up to being a "physicist."

While I may have understood all of the equation from a mathematical and physical standpoint back when I was doing pure physics, I fear that my knowledge that would allow me to easily pick this up has since degraded. So my final questions would be:

1. After being solely focused on clinical medical physics the past few years, having lost a lot of math's and physics knowledge, do you have any recommendations as how I could get up to speed with E&M, particle physics and the math's necessary for this? I have been studying neural networks in my spare time, is it just the case of picking up old textbooks? Does the Canadian medical physics system prepare you for this level of pure physics involved in medical physics? Are there any resources?

2. How can I actually participate in topics like this. My hospital has an associated university, and the university even have an MRI scanner here, but we are not a "university hospital" per say. I have briefly met the medical physics academic staff, but after looking into all of their research one-by-one, it consists of assisting medical doctors with obtaining clinical outcomes from clinical trials, exploring biomarkers for clinical outcomes and some that just seem like pure clinical work such as investigating brain plasticity, drug development etc. (so not even sure how that's medical physics never mind anything to do with physics?).

 

[Choppy]

1. After being solely focused on clinical medical physics the past few years, having lost a lot of math's and physics knowledge, do you have any recommendations as how I could get up to speed with E&M, particle physics and the math's necessary for this? I have been studying neural networks in my spare time, is it just the case of picking up old textbooks? Does the Canadian medical physics system prepare you for this level of pure physics involved in medical physics? Are there any resources?

One of the challenges with "picking up a textbook and working through it" is that most people will do it for a little while, but without deadlines or specific goals it tends to lose out to other demands that have them. Six months down the road, you're bookmarked in chapter 3 with a plan to get back to in "eventually." (For what it's worth, there are a few rare people who have the focus and capacity to accomplish this though.) What tends to work for me is project-based (re-)learning. If you studied the concepts and understood them once, having a project that draws on that knowledge base will require you to look up what you need. You work through simplified versions of your project until you solve them and then add more complexity until you have something that does what you need it to do. There can be a lot of testing and validation involved, so it's not exactly efficient. Other ideas might include taking a refresher course or finding a workshop... depending on what specifically it is you want to refresh.

2. How can I actually participate in topics like this. My hospital has an associated university, and the university even have an MRI scanner here, but we are not a "university hospital" per say. I have briefly met the medical physics academic staff, but after looking into all of their research one-by-one, it consists of assisting medical doctors with obtaining clinical outcomes from clinical trials, exploring biomarkers for clinical outcomes and some that just seem like pure clinical work such as investigating brain plasticity, drug development etc. (so not even sure how that's medical physics never mind anything to do with physics?).

Often this is a matter of building up your scientific network. How many of these people have you talked to directly? Just because someone publishes in one area, doesn't mean that's their only interest. It could be that they also want to do something more physics-related, but the funding opportunities available to them have been more medical in nature. If you can generate opportunities to talk to them, brainstorm, and think about potential projects, not all of them will pan out into something productive, but some might.
You can expand your network through conferences, workshops, courses, etc. You can reach out to local universities and offer to give presentations on medical physics. I don't know if the UK has anything equivalent to COMP or AAPM, but I've found that volunteering for such groups really helps to build up your network, even if you're only volunteering on the local arrangements committee for a conference.

 

[Quantum_Cthulhu]

One of the challenges with "picking up a textbook and working through it" is that most people will do it for a little while, but without deadlines or specific goals it tends to lose out to other demands that have them. Six months down the road, you're bookmarked in chapter 3 with a plan to get back to in "eventually." (For what it's worth, there are a few rare people who have the focus and capacity to accomplish this though.) What tends to work for me is project-based (re-)learning. If you studied the concepts and understood them once, having a project that draws on that knowledge base will require you to look up what you need. You work through simplified versions of your project until you solve them and then add more complexity until you have something that does what you need it to do. There can be a lot of testing and validation involved, so it's not exactly efficient. Other ideas might include taking a refresher course or finding a workshop... depending on what specifically it is you want to refresh.

Yeah this sounds practical - I have found myself, for the majority, only re-learning things when a specific project calls for it.

Often this is a matter of building up your scientific network. How many of these people have you talked to directly? Just because someone publishes in one area, doesn't mean that's their only interest. It could be that they also want to do something more physics-related, but the funding opportunities available to them have been more medical in nature. If you can generate opportunities to talk to them, brainstorm, and think about potential projects, not all of them will pan out into something productive, but some might.

I have spoken to the majority of them personally, but not directly about physics related medical physics. My only concern would be how to approach it. Solving the LBTE in the presence of magnetic fields is all well if there is, or is going to be a MR-Linac which is not going to be present within the nearest 100 miles from me for a long time.

I think the approach you outlined about talking to them and brainstorming is sensible, but I'm not sure how much time a professor is going to allocate to a non-academic to brainstorm without some foundation of a project. I can always try though.

You can expand your network through conferences, workshops, courses, etc. You can reach out to local universities and offer to give presentations on medical physics. I don't know if the UK has anything equivalent to COMP or AAPM, but I've found that volunteering for such groups really helps to build up your network, even if you're only volunteering on the local arrangements committee for a conference.

Our COMP/AAPM equivalent is IPEM. I have attended many conferences particularly in MRI and listened to over one hundred talks by now, but again all mostly QA/assisting clinical research/service related. I suppose I could do as you say and take a more active role in IPEM with the hope of finding a like-minded person through networking.

Thank you for your responses.

 

[Gruxg]

I suppose my frustration comes from the title itself. I just want to be expected to do some physics. Even if it's directly relevant. Like some particle physics in relation to radiotherapy. Some EM related to MRI, nuclear physics related to nuclear medicine etc.

You are not alone in this feeling. The clinical job does not have too much physics actually. I don't know in Canada, but in most European countries it is currently more like a mixture of project manager and quality engineer (and dosimetrist, if you are in the radiotherapy field). The physics we use was well established decades ago, nowadays it is mostly encapsulated into black boxes by the different vendors, but the technology, especially the IT, is increasingly more complex, so the field evolved from a physics branch to a rather technical, interdisciplinary (and necessarily very pragmatic) profession. And I think it is probably time to accept it and stop marketing the job as if it were applied physics. In some way it is like chemistry vs. pharmacy (the later is based on the former to a large extent, but it is a different, independent field).

The ransomware hazards in hospitals and the draconian restrictions imposed in the last years by the system administrators "for safety reasons" don't help either, because some weeks you spend like 40% of the time fighting with the IT issues derived from these restrictions or trying to get the tools you need to do your work, so, forget about using Matlab or even Python or MC codes in the hospital computers.