Re: [meddos] Re: Pinnacle and Varian
Pinnacle List Server
"Again all in my humble opinion…alright, let the ways in which this is wrong now be debated...", I 'll just add my comments in red after each of yours...
Roy is correct that Rapid Arc in name is exclusive to Varian—but VMAT, as a technology, is not.
Pinnacle has SmartArc which is also capable of utilizing the dynamic Dose Rate modulation available on Varian accelerators.
Here are seven things I like about Pinnacle: (More accurately "7 Things I hate about Eclipse)
1. Ease with which to manage patient database (try to create a copy of patient in Eclipse—cannot do it) Managing a patient database and creating a copy of a patient, are two entirely different functions. Managing the patient database in Eclipse is very easy as it pertains to "Backup and Restore", nightly auto-backups for disaster recovery, and purging records after backed up to a different server for storage. The inability to "Copy a patient" is an intentional limitation for safety reasons. A duplicate copy of a patient's information, to me, would seem confusing and I can't think of a reason to have one when it is simple enough to and paste a plan into a different course or even re-read the ct dataset into a "test" patient and even copy and paste the plan with standard windows functions.
2. Ease of understanding by users the simple function of normalizing a plan (Pinnacle wins BY FAR over Eclipse’s Plan and Prescribed normalization functions) In current versions of Eclipse when you click on "Normalization" you are presented with a straight forward list of options including "100% at Isocenter, ___% of dose to cover ___% of structure, ___% Normalization, No Plan Normalization" and many more. Where's the confusion or lack of ease of understanding???
3. Want to run a plan using multiple Rx’s like: SCF, Axilla and Breast: Giving the nod to Pinnacle here as the only to do this in Eclipse is via a function called Plan Sum (which is not a Plan at all but really an evaluation tool)…and a real pain (IMHO) Isn't the entire Treatment Planning System an evaluation tool??? Plan Sum is a powerful tool and saying powerful doesn't even do it justice. Not only can you take two plans that were performed on two different ct datasets and combine them into a composite plan, you can actually create new plans from within Plan Sum utilizing as many plans and different prescriptions as you want. This can be done in hindsight as well. You don't have to start with and build from a composite plan and then separate the individual plans into different trials... Just add them all together when you're done. How is that a pain???
4. Scripting—Maybe a upgrade of C+ coming but the limitations of .tml based templates in Eclipse is horrible. Scripting. Are you referring to the ability to "record" your button clicks in Pinnacle and saving it as a script? And doing so, if you make a mistake you have to start from the beginning to record it all over? And then later when you use the "script" since it is patient specific when you use it and something doesn't match it crashes? Or if you're bright enough to know all the UNIX commands you can edit it but finding the script is buried deep in the bowels of Pinnacle. Well you're right, Eclipse now supports "Scripting" and it comes built in with a variety of built in scripts that can perform functions like retrieving center coordinates for structure location, reporting structure volume at a specific dose and locating it on the DVH for you, etc. It gathers the information from the patient database and supports all ,NET languages, not just a single language, v11 supports C#. With a knowledge of any of these languages, very powerful scripts can be written just as easily as can be written in Pinnacle, which only utilizes UNIX.
5. Output of basic plan text info.—Quick, try to tell me the dose to a point and the effective depth in a plan run with Eclipse—I’ll be over here putting in my .pdf from Pinnacle with all that information already highlighted and in the R&V for therapists. Depends on which text template you choose to display what information you desire to be displayed, but the Full Template, though it may display a lot of information, displays exactly the information you just referenced, dose to every reference point, SSD to each point, measured depth to each point, effective depth to each point... and as I'm sure it is with Pinnacle, you just have to know where to look to read that information. And, it can be printed to PDF into Mosaiq or live as a live "read only" plan in ARIA... with scrolling through isodose lines and everything. Again, I don't understand where the confusion is coming from.
6. MLC editing. Pinnacle wins here. Have you ever wanted to create a second block maybe to block out a hot spot—try “adding” a second block in Eclipse—nice try…have to re-contour the whole bloody original block. Again, here you state two different things. MLC editing and adding a second block. MLC editing in Eclipse is simple. You can edit a MLC leaf one at a time, by an entire bank of MLCs, or by using your shift key, select a section of MLC's to move. You can add second blocks all day long in Eclipse, and third, and fourth, if you're using blocks. And they all have the same priority, They all block beam. You do not have to assign a priority to them. Now if you want to redraw the area covered by an MLC, you are right, you have to redraw the entire area. But to just block out a hotspot, why not just drag the MLC's that need to be, over the hotspot? Additionally, if you had utilized the irregular surface compensator feature whbuilds a fluence over the intended area to treat based on patient shape, depth, and the heterogeneous properties of the patient in the field, you could simply paint the hotspot away by editing the fluence.
7. Add a wedge and hope you get it right…Pinnacle at least shows one the little picture of the wedge as you rotate the heel…all you are presented with in Eclipse is the laundry list of every available wedge and heel orientation in user-friendly jargon like: 15IN30OUT…wait was that a 15 wedge or 30 wedge?? Again, where is the confusion?? The drop down list plainly describes all of the wedges available. For example, W15IN30, W15OUT30, W15L30, W15R30... These are all physical wedges and it is obvious if you know anything about the physical characteristics of your wedges these are all 15W that cover a 30cm wide field. As you get to the 60W, they read W60IN15 and so forth. If you select a wedge and click "APPLY" it displays the orientation of the wedge on the patient. If it's not right, pick the opposite. But who uses physical wedges anymore?? The dynamic wedges are listed as "EDW10OUT, EDW15IN, EDW15OUT, EDW30R, etc...
BONUS: Completely too many crashes/unusable errors encountered during planning—Quick what does “::iosfailbitset” error mean to me? Not saying that Pinnacle doesn’t crash, but at least it is infrequent (plug for UNIX) and I don’t typically get kicked out for reasons totally foreign to me. Again, what version are you talking about??? Since the release of version 10 two years ago, Eclipse is now on a 64bit architecture and is extremely stable. The crashes you are referring to were simply inherent to the issues with older versions of windows. And even now, moving to v11, Eclipse is moving to a SQL database which is even more stable. And even though Pinnacle may not crash often, as you claim, it does as well. And as I have read on this list serve, with the release of 9.2, it is slower to load than with 9.0 and crashes many times over than 9.0, and these are statements from the ones using 9.2, not me, as I haven't used 9.2 clinically yet.
Trust me, I could elaborate on 10 other things, but let’s keep it simple. So bring on the other 10 you can easily elaborate on rather than keeping it simple. How about deformable registration, Knowledge-based autocontouring with built in filters based on staging, Monte Carlo equivalent photon algorithms, eMC (I know you state below it's not needed but Pinnacle users have been screaming for it for years and Eclipse has had it), Proton planning and some things I touched on like combining plans from different CT datasets, irregular surface compenstaors and fluence editing. This is all functionality Eclipse currently has integrated into the system that Pinnacle can't touch. Since this is a Pinnacle User Group, I didn't really expect to see much support for Eclipse, but since you opened it up for debate, I thought I would elaborate on a defense for Eclipse
Here are seven things I like about Eclipse:
1. Ease with which to verify/analyze DICOM objects. Pinnacle: Add in a non-DICOM study (even though it contains a .dcm tag) and access the Images folder through the Planning GUI. They just do not appear in the list yet the images are in /autoDataSets/DICOM folder. Don’t know what’s wrong with them: Pinnacle—too bad, Eclipse: Dicom2Ascii and look it up. Have had several instances of images not being seen in the DICOM directory, open them up in Eclipse, re-export to a shared directory, manually place into /autoDataSets/DICOM directory and magically they now appear..what is that about?
2. Color options—sounds picky but the Pinnacle really is awful here. What is Pinnacle thinking with a color called Lavender…that you cannot see what the item is that is shaded this color. I equate it to having the most important item of this email written here in white …did you see it?—Thank you Varian for the seemingly endless supply of color options.
3. Contouring—Oh man, some of the tools in Eclipse (Density based brushed tool for example) simply awesome—c’mon Pinnacle…need to ante up here.
4. Copy and Paste of plans. I mean is ANYONE not familiar with this basic concept. Now take what you know and try to do this in Pinnacle onto a separately fused data set---wake me when you figure out a way to easily do this (yes, there are script workarounds…but why should there be a workaround for this most necessary and all too often occurring clinical situation)
5. NTO – this is nice (assuming one knows how to use it)…and man, I hate all of those rings in Pinnacle.
6. e-MC – not really sure it’s necessary, but when you have the “gold standard” in your suite of algorithms, shouldn’t you get the nod.
7. If you have Aria—no comparison as you are essentially planning in Aria and the risk of data being compromised through a transfer is negated.
BONUS: Helpdesk ability to remotely access my computer and walk through a situation with me, where I can watch what they are doing and how. NOT AT ALL into the Pinnacle manner of helpdesk assistance.
Again all in my humble opinion…alright, let the ways in which this is wrong now be debated…
-Mark
.
__._,_.___![Yahoo! Groups]()
.![]()
__,_._,___