Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Feasibility of Dynamic SPECT-Renography with Automated Evaluation Using a Deep Neural Network

Rogowski, Viktor LU (2021) MSFT01 20202
Medical Physics Programme
Abstract
Introduction: Renography is a standard diagnostic examination that evaluates renal function, renal pelvic dilatation and urinary obstruction. Renography is performed by injecting a radiopharmaceutical (predominately 99mTc-MAG3) and using gamma camera to image the biodistribution in a dynamic sequence. Evaluating renography images encompasses many difficulties regarding background correction for activity in blood and tissue located anterior and posterior of the kidney. With obstructed renal function activity accumulates in the liver and border of the kidneys is difficult to distinguish. A possible solution to these difficulties is the novel multi-detector Single Photon Emission Computed Tomography (SPECT) systems which, unlike conventional... (More)
Introduction: Renography is a standard diagnostic examination that evaluates renal function, renal pelvic dilatation and urinary obstruction. Renography is performed by injecting a radiopharmaceutical (predominately 99mTc-MAG3) and using gamma camera to image the biodistribution in a dynamic sequence. Evaluating renography images encompasses many difficulties regarding background correction for activity in blood and tissue located anterior and posterior of the kidney. With obstructed renal function activity accumulates in the liver and border of the kidneys is difficult to distinguish. A possible solution to these difficulties is the novel multi-detector Single Photon Emission Computed Tomography (SPECT) systems which, unlike conventional systems with which only a few projections are measured simultaneously, allows for a simultaneous 360-degree measurement of the biodistribution and thereby enables dynamic SPECT. However, with the new system, other problems arise regarding image noise and larger workload on the technologist with evaluation. By using convolutional neural network (CNN) to do automatic segmentation these obstacles can be avoided. However, to train a CNN thousands of images are required which can be obtained with Monte Carlo simulations.

The aim of this thesis is to evaluate the feasibility of four dimensional (4D)-renography and compare it to conventional renography and to train a CNN for three dimensional (3D) semantic segmentation of 4D-renography images with the use of Monte Carlo simulated images and to evaluate the model.

Material and Methods: 15 4D anthropomorphic digital phantoms (XCAT) with 15 different sets of renal functions and split renal functions were simulated with the Monte Carlo program SIMIND to mimic the VERITON Cadmium Zinc Telluride (CZT) camera. Simulations were set to 360-degree rotation and 120 projections. Each organ was simulated individually, and ground truth masks of the organs were created from the phantoms. The simulations were assembled into 3 sets of data: two dimensional (2D) planar images with posterior projection, which resemble the images corresponding to conventional renography studies, geometric mean corrected images and 3D dynamic tomographic projections. The assembled images were normalized to a realistic activity level, acquisition time, sensitivity correction and Poisson distributed noise was then added. Tomographic projections were further reconstructed using an in-house OS-EM method. Evaluation was performed with the use of simulated masks and with extracted count rate a split renal function is calculated and compared between the different sets.
A CNN was trained by using 3D U-net network. Tomographic images were modified resulting in 24894 unique training images and masks. Evaluation of split renal function using CNN segmentation was compared to planar evaluation.

Results: The tomographic evaluation showed less deviation from the true value in the majority of cases when compared to posterior planar evaluation. For lower renal function with greater difference in split renal function planar evaluation showed greater deviation from true value. Geometric mean image evaluation showed greater deviation for low renal function and generally worse than both tomographic and posterior planar evaluation. Evaluation using neural network showed less deviation for most cases compared to posterior planar evaluation.

Conclusion: The results show that 4D-renography may be feasible, with a possible accuracy of a split renal function measurement that is better than the conventional methods. Scaling the collimator sensitivity to the VERITON cameras sensitivity results in noisy images which limits the minimum possible voxel size and time resolution. Evaluation with the CNN shows great promise. Further development and evaluation of clinical images is needed to ensure accuracy. (Less)
Popular Abstract (Swedish)
Nuklearmedicin är en inriktning inom radiologi där väldigt små mängder av radioaktiva ämnen används för diagnostik av organfunktioner. Inom diagnostik är förbättring av bildtagningssystem ständigt eftertraktade. Ett sätt att utveckla och förbättra systemen är genom simulering. Med utveckling av artificiell intelligens (AI), har möjligheten för utvärderingar av nuklearmedicinska bilder uppstått med hjälp av AI.

Renografi är en diagnostisk undersökning för att utvärdera njurfunktion, utvidgning av njurbäcken och urinvägsobstruktion. Renografi utförs genom att injicera ett radiofarmaceutiskt läkemedel (huvudsakligen 99mTc-MAG3) som sedan detekteras av en gammakamera för att avbilda biodistributionen i en dynamisk sekvens. Vid utvärdering... (More)
Nuklearmedicin är en inriktning inom radiologi där väldigt små mängder av radioaktiva ämnen används för diagnostik av organfunktioner. Inom diagnostik är förbättring av bildtagningssystem ständigt eftertraktade. Ett sätt att utveckla och förbättra systemen är genom simulering. Med utveckling av artificiell intelligens (AI), har möjligheten för utvärderingar av nuklearmedicinska bilder uppstått med hjälp av AI.

Renografi är en diagnostisk undersökning för att utvärdera njurfunktion, utvidgning av njurbäcken och urinvägsobstruktion. Renografi utförs genom att injicera ett radiofarmaceutiskt läkemedel (huvudsakligen 99mTc-MAG3) som sedan detekteras av en gammakamera för att avbilda biodistributionen i en dynamisk sekvens. Vid utvärdering av renografibilder uppstår svårigheter såsom bakgrundskorrigering för aktivitet i blod och vävnad som omringar njuren. Med nedsatt njurfunktion ackumuleras aktivitet i levern och omkringliggande blodfylld vävnad vilket bidrar till att njurgränsen blir svår att urskilja och kvantifiering av njur sidofördelningen blir mer osäker. En möjlig lösning för dessa svårigheter är de nya multi-detektor SPECT systemen som, till skillnad från konventionella system med vilka endast ett fåtal projektioner mäts samtidigt, möjliggör en 360-graders mätning av biofördelningen samtidigt och dynamisk SPECT. Med det nya systemet ökar problem med bildbrus och större arbetsbelastning för teknologen med utvärdering av bilder. Genom att använda konvolutions neurala nätverk (CNN) som är ett typ av AI, för att göra automatisk segmentering kan dessa hinder undvikas. Men för att träna ett CNN krävs det tusentals bilder som kan erhållas med hjälp av Monte Carlo-simuleringar.
Syftet med detta arbete är att utvärdera genomförbarheten av 4D-renografi och jämföra den med konventionell renografi samt att undersöka om det är möjligt att träna ett CNN för 3D-semantisk segmentering av 4D-renografibilder som är tillräckligt bra för att i framtiden kunna utgöra ett verktyg vid utvärdering av kliniska undersökningar.

Genom att simulera mänskliga digitala fantom med olika uppsättningar av njurfunktioner och delade njurfunktioner kan man efterlikna en renografiundersökning med olika kamerasystem. Simulationen genomförs med Monte Carlo baserade programmet SIMIND som baseras på genererade likformigt fördelade slumptal för modellering av strålningsprocesser. I SIMIND definieras kamerainställningar, fantomsinställningar och radioaktiva källans specifikationer. Från simuleringarna erhålls 2D bilder efterliknande kliniskt använda kamerasystemet och 3D bilder efterliknande nya kamerasystemet. Ett CNN tränas med hjälp av 3D U-net-nätverk.

Resultatet i det här arbetet visar att utvärdering med användning av det nya SPECT systemet är bättre än det konventionella. Men detta medför dock en bekostnad av ökat bildbrus som begränsar minsta möjliga voxelstorlek och tidsupplösning. Utvärdering av 4D-renografibilder med användning av AI visar goda resultat samt en möjlighet för framtida användning. Ytterligare utveckling och utvärdering av kliniska bilder behövs för att säkerställa noggrannheten. (Less)
Please use this url to cite or link to this publication:
author
Rogowski, Viktor LU
supervisor
organization
course
MSFT01 20202
year
type
H2 - Master's Degree (Two Years)
subject
language
English
id
9050807
date added to LUP
2021-06-08 07:57:37
date last changed
2021-06-08 08:11:47
@misc{9050807,
  abstract     = {{Introduction: Renography is a standard diagnostic examination that evaluates renal function, renal pelvic dilatation and urinary obstruction. Renography is performed by injecting a radiopharmaceutical (predominately 99mTc-MAG3) and using gamma camera to image the biodistribution in a dynamic sequence. Evaluating renography images encompasses many difficulties regarding background correction for activity in blood and tissue located anterior and posterior of the kidney. With obstructed renal function activity accumulates in the liver and border of the kidneys is difficult to distinguish. A possible solution to these difficulties is the novel multi-detector Single Photon Emission Computed Tomography (SPECT) systems which, unlike conventional systems with which only a few projections are measured simultaneously, allows for a simultaneous 360-degree measurement of the biodistribution and thereby enables dynamic SPECT. However, with the new system, other problems arise regarding image noise and larger workload on the technologist with evaluation. By using convolutional neural network (CNN) to do automatic segmentation these obstacles can be avoided. However, to train a CNN thousands of images are required which can be obtained with Monte Carlo simulations. 

The aim of this thesis is to evaluate the feasibility of four dimensional (4D)-renography and compare it to conventional renography and to train a CNN for three dimensional (3D) semantic segmentation of 4D-renography images with the use of Monte Carlo simulated images and to evaluate the model.

Material and Methods: 15 4D anthropomorphic digital phantoms (XCAT) with 15 different sets of renal functions and split renal functions were simulated with the Monte Carlo program SIMIND to mimic the VERITON Cadmium Zinc Telluride (CZT) camera. Simulations were set to 360-degree rotation and 120 projections. Each organ was simulated individually, and ground truth masks of the organs were created from the phantoms. The simulations were assembled into 3 sets of data: two dimensional (2D) planar images with posterior projection, which resemble the images corresponding to conventional renography studies, geometric mean corrected images and 3D dynamic tomographic projections. The assembled images were normalized to a realistic activity level, acquisition time, sensitivity correction and Poisson distributed noise was then added. Tomographic projections were further reconstructed using an in-house OS-EM method. Evaluation was performed with the use of simulated masks and with extracted count rate a split renal function is calculated and compared between the different sets.
A CNN was trained by using 3D U-net network. Tomographic images were modified resulting in 24894 unique training images and masks. Evaluation of split renal function using CNN segmentation was compared to planar evaluation.

Results: The tomographic evaluation showed less deviation from the true value in the majority of cases when compared to posterior planar evaluation. For lower renal function with greater difference in split renal function planar evaluation showed greater deviation from true value. Geometric mean image evaluation showed greater deviation for low renal function and generally worse than both tomographic and posterior planar evaluation. Evaluation using neural network showed less deviation for most cases compared to posterior planar evaluation.

Conclusion: The results show that 4D-renography may be feasible, with a possible accuracy of a split renal function measurement that is better than the conventional methods. Scaling the collimator sensitivity to the VERITON cameras sensitivity results in noisy images which limits the minimum possible voxel size and time resolution. Evaluation with the CNN shows great promise. Further development and evaluation of clinical images is needed to ensure accuracy.}},
  author       = {{Rogowski, Viktor}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Feasibility of Dynamic SPECT-Renography with Automated Evaluation Using a Deep Neural Network}},
  year         = {{2021}},
}