![Show Menu](styles/mobile-menu.png)
![Page Background](./../common/page-substrates/page0456.jpg)
S441
ESTRO 36
_______________________________________________________________________________________________
3
University Hospital Heidelberg, Radiation Oncology,
Heidelberg, Germany
Purpose or Objective
To develop and evaluate a new concept for automatic re-
planning of VMAT plans as failure concept for solitary
treatment machines, e.g. MR-Linac. In contrast to
previously published automatic planning approaches
which replicate the planned dose distribution, we propose
an automatic re-planning concept which uses constrained
optimization to generate Pareto-optimal VMAT plans for
different treatment machines. The scheme interprets a
treatment plan as a point on the corresponding Pareto
front, and creates the re-planned one by projecting this
point onto the substitute´s Pareto front. Thereby,
comparable biological effect and hence clinical outcome
can be guaranteed.
Material and Methods
In this automatic re-planning study, n=16 prostate cancer
and n=19 head and neck cancer (HNC) cases were
included. All patients had previously planned clinical
VMAT plans created with in-house TPS Hyperion. Hyperion
uses constrained optimization where a Lagrange multiplier
λi is associated to each cost-function constraint Ci, rating
the effect of each organ-at-risk (OAR) constraint on the
target objective.
Automatic re-planning starts from the initially reached
optimal constraints Ci for PTVs and OARs and adapted
machine parameters. A full optimization was executed
automatically, in order to generate a comparable Pareto-
optimal plan. For prostate cases, Elekta BeamModulator
plans were re-planned for Elekta Agility, whereas for HNC,
Elekta Agility plans were re-planned for Elekta MLCi.
For prostate cases we identified rectum and bladder as
main OARs and for HNC contralateral parotid gland and
spinal cord. For PTV we evaluated variations in EUD, D
Mean
,
D
2%
and D
98%
and for OARs EUD and D
2%
.
Results
Automatic re-planning using constrained optimization was
successful in all cases. Auto-optimized plans never
corrupted OAR constraints, in some cases re-planning even
improved OAR sparing. The mean deviation (range) in
rectum EUD was 0,30% (-1,04 – -0,27%), bladder EUD 0,44%
(-1,08 – -0,13%), parotid EUD -0,34% (-14,79 – 8,23%) and
spinal cord EUD -0,02% (-0,49 – 0,31%). For the prostate
cases the mean EUD deviation in PTV was -0,15% (-0,57 –
0,56%) and for the HNC cases -0,60% for PTV_60 (-2,58
– -0,08%) and -0,79% (-3,44 – 0,20%) for PTV_54,
respectively. Except of 3 HNC cases, all evaluated
parameters for targets showed variations within ±1%. For
3 HN cases the target EUD is reduced by up to 3.44%,
indicated by λ > 10 * λ
avg
. Consequently, if all λ < 10* λ
avg
,
the original and the re-planned plan comply with the given
constraints and therefore represent the same optimal
point on the Pareto-front, which means they are equal in
terms of biological effect for targets and OAR.
Conclusion
This study showed that fully automatic re-planning by
taking a prescription list from previously optimized VMAT
plans is feasible and successful in terms of equal plan
quality. Furthermore this approach enables the
identification of problematic plans beforehand.
PO-0822 An evolutionary model improvement strategy
for knowledge-based planning
Y. Zhang
1
, F. Jiang
1
, H. Yue
1
, S. Li
1
, Q. Hu
1
, M. Wang
2
, H.
Wu
1
1
Key Laboratory of Carcinogenesis and Translational
Research Ministry of Education/Beijing- Department of
Radiation Oncology- Peking University Cancer Hospital &
Institute, Department of Radiation Oncology, Beijing,
China
2
National Institute for Radiological Protection- China
CDC, National Institute for Radiological Protection-,
Beijing, China
Purpose or Objective
It was reported that RapidPlan, a knowledge-based
solution, can improve planning efficiency, quality and
consistency. Should the performance of RapidPlan be
dependent on the quality of the plans training the model,
this study hypothesizes that RapidPlan can improve its
constituent plans (closed-loop) and improve the model
itself by incorporating the better knowledge (evolution).
Moreover, the maximum number of iterations to exhaust
the full potential was also tested.
Material and Methods
An initial RapidPlan model (M
0
) for pre-surgical rectal
cancer patients was configured using 81 best-effort
manual VMAT plans (P
0
) with SIB (50.6 Gy and 41.8 Gy to
95% PTV
boost
and PTV respectively). For simplification,
decreased or increased mean dose to both femoral head
(FH) and urinary bladder (UB) were considered as
improved (P
+
) or worsened plans (P
-
) respectively. P
±
denoted intertwined plans. The first closed-loop iteration
of re-optimizing P
0
using M
0
yielded P
1
: 69 P
1+
, 12 P
1±
and
0 P
1-
. By substituting P
1+
for their corresponding P
0
, the
library of model M
1+
consisted of 12 P
0
and 69 P
1+
. The
second closed-loop iteration of re-optimizing P
0
using M
1+
produced 35 P
2+
that were superior to both P
0
and P
1
,
hence the knowledge base of M
2+
composed 9 P
0
, 37 P
1+
and 35 P
2+
. As open-loop validation, 30 clinical plans (P
v
)
that were not included in the model were re-optimized
using each model. Re-optimization maintained all
parameters except the MLC sequences were redesigned
using the objectives generated by the models.
Renormalizations to target prescriptions were performed
to make OAR dose comparable.
Results
Consistent with literature, knowledge-based so lution
improved the plan quality in both closed- and open-loop
validations than the conventional trial-and-error process.
In the first closed-loop evolution, the mean±SD of D
mean_FH
and D
mean_UB
for 69 P
1+
were 12.88±1.38 Gy and 23.06±3.11
Gy respectively, which were significantly lower by 23.70%
and 9.53% than the corresponding values of P
0
(P<0.05). In
the second round of closed-loop re-optimization, the
D
mean_FH
of the 35 P
2+
decreased to 11.12±1.48 Gy
(corresponding P
0
: 17.13±2.06 Gy; P
1
: 13.08±1.36 Gy), and
D
mean_UB
decreased to 22.80±3.72 Gy (corresponding P
0
:
25.79±3.34 Gy; P
1
: 23.41±3.59 Gy). Table 1 and figure 1
display the open-loop validation results for various
models. The marginal disparities of HI and CI
(magnitudes
≦
0.04) and largely overlapped DVH lines
indicated comparable target dose distribution. In line with
the closed-loop test, RapidPlan reduced the dose to OARs
massively than clinical plans in the open-loop validation
(Fig 1). The first model evolution has greatly and
significantly lowered the dose to FH at comparable dose
to the UB and targets. The second iteration has made little
difference though.