[Gate-users] Dose Uncertainty

[Mahdjoub Hamdi]

Hi Andrew,
Sorry for the delay. The paper title is: Reporting and analyzing statistical uncertainties in Monte Carlo–based treatment planning

Thanks,
Mahdjoub

From: Nixon Andrew (RBV) NHS Christie Tr <Andrew.Nixon at christie.nhs.uk> 
Sent: Wednesday, September 04, 2019 2:24 AM
To: Mahdjoub Hamdi <mahdjoub.hamdi at hotmail.fr>
Subject: RE: Dose Uncertainty

Hi Mahdjoub,

Thanks for sharing this. Can you give me the full reference for the Chetty paper please? I don’t think I’ve read it.

Kind regards,

Andrew.

From: Gate-users [mailto:gate-users-bounces at lists.opengatecollaboration.org] On Behalf Of Mahdjoub Hamdi
Sent: 04 September 2019 02:18
To: meryem fiak; Mahdjoub Hamdi
Cc: mailto:gate-users at lists.opengatecollaboration.org
Subject: Re: [Gate-users] Dose Uncertainty

Hello Meriem,

Sorry for the delay I lost my script somewhere. Here is how I calculated the statistical uncertainty, you can adapt it to read multiple files on the cluster, you have to put your number of primary emitted particles:


% This program calculates the statistical uncertainty for Monte Carlo
% simulations and plot the DVH for the dose and statistical uncertainties.

% Mahdjoub Hamdi, P.h.D.
% Washington University in Saint Louis.

N=180000000;  % Number of primary photons

d1=analyze75read('000000_main-Dose.hdr');   % Read the dose matrix
u1=analyze75read('000000_main-Dose-Squared.hdr'); % Read the uncertainty matrix

 
sdk1 = sqrt(1/(N-1).*(((u1)./N)-((d1/N).^2)));  % equation 2 Chetty et al   
 
d2=analyze75read('000090_main-Dose.hdr');
u2=analyze75read('000090_main-Dose-Squared.hdr');

sdk2 = sqrt(1/(N-1).*(((u2)./N)-((d2/N).^2))); % equation 2 Chetty et al  
 
d3=analyze75read('000315_main-Dose.hdr');
u3=analyze75read('000315_main-Dose-Squared.hdr');

sdk3 = sqrt(1/(N-1).*(((u3)./N)-((d3/N).^2))); % equation 2 Chetty et al

sdk=sqrt((1/3.*sdk1)+(1/3.*sdk2)+(1/3.*sdk3)); 
  
d = d1+d2+d3;
err=100*(sdk./d); % statistical uncertainty


Good luck,
Mahdjoub




From: meryem fiak <mailto:fiakmeryem.13 at gmail.com> 
Sent: Wednesday, August 28, 2019 9:21 AM
To: mailto:mahdjoub.hamdi at hotmail.fr
Subject: Dose Uncertainty

Bonjour,

Je suis doctorante en physique nucléaire, mon travail porte sur la simulation des linac en radiothérapie à l'aide du code MC GATE. J'ai calculé les fontions dosimétriques du linac opérant à 18 MeV (PDD et Profiles) en laçant plusieurs jobs et je les merge vers la fin. (Output: Root)
Je rencontre un problème au niveau du calcul de l'incertitude statistique sur la dose. Je voudrais savoir qu'est ce que dois je faire afin d'obtenir l'incertitude sur la totalité des jobs. A titre d'exemple, j'ai lancé 1.4E11 particules primaires réparties sur 140 jobs. 
J'ai consulté pas mal d'articles qui recommendent en cas de calcul en parallèl de calculer l'incertitude statistique relative pour chaque job d'une manière retrospective à partir de la formule proposé par [chetty,2006], chose que j'ai pas su comment l'appliquer.


Je vous saurais gré de bien vouloir m'apporter de l'aide dans ce sens.

Cordialement.



-- 
Ms.FIAK Meriem
-------------------------------------------------------------------
Job Title: Phd Student 
--------------------------------------------------------------------
Hassan II University of Casablanca
Faculty of sciences Ben M'sik
Department of physics 
Laboratory of the Physics of Condensed Matter (LPMC)
Team of Subatomic and Applications Research (ERSA)
---------------------------------------------------------------------
E-mail: mailto:fiakmeryem.13 at gmail.com