Table of Contents Table of Contents
Previous Page  98 / 146 Next Page
Information
Show Menu
Previous Page 98 / 146 Next Page
Page Background

Introduction:

AAPM TaskGroup204 introduced size-specific dose esti-

mates (SSDE) for adult and pediatric patients undergoing body CT

examinations. This investigation aims to extend that work to overweight

and obese patients by using simulations of clinical data on hybrid body mass

index (BMI)-adjustable phantoms.

Methods and Materials:

Dosimetric data of 335 patients who underwent

abdominal CT scans with two different scanners were extracted from

NexoDose™ database (Bracco Diagnostics, Inc). Patients’ waist effective di-

ameter (ED) was calculated according to AAPM204. Patients were divided

in four weight-classes matching the hybrid RPI BMI-adjustable phan-

toms: normal weight, overweight and obese (grade I and II). For each weight-

class mean scan parameters were calculated and simulated using

VirtualDose™ CT (VirtualPhantoms, Inc.). For each weight-class and totally

included organs, scanner-independent CTDIvol-to-organ-dose conversion

coefficients were calculated by normalizing organ doses by scanner-

specific 32 cm CTDIvol values and averaging across the scanners. Correlation

between conversion coefficients and ED was explored.

Results:

A strong correlation between CTDIvol and ED was observed for

all abdominal protocols. On the contrary an inverse correlation between

CTDIvol normalized organ dose and ED was observed. CTDIvol-to-organ-

dose conversion coefficients are smaller than the coefficients defined by

AAPM TG204 for the same ED. This is inherent to the difference between

phantoms used by AAPM204 and hybrid phantoms considered in this study,

modeled with different percentages of subcutaneous adipose tissue ac-

cording to patient’s BMI.

Conclusions:

For all abdominal scans that have been analyzed, strong cor-

relation exists between CTDIvol normalized organ dose and ED. Results

indicate that robust estimates of BMI-specific organ doses may be provid-

ed using the size-specific, scanner-independent CTDIvol-to-organ-dose

conversion coefficients described in this work.

http://dx.doi.org/10.1016/j.ejmp.2016.01.319

B.315

CT IMAGE QUALITY ASSESSMENT BY A CHANNELIZED HOTELLING

OBSERVER (CHO): APPLICATION TO PROTOCOL OPTIMIZATION

A. Taddeucci

* , a ,

L. Noferini

a ,

M. Bartolini

b ,

I. Mench

i b .

a

Health Physics Unit,

Azienda Ospedaliero-Universitaria Careggi, Firenze, Italy;

b

Medical Imaging

Department, Azienda Ospedaliero-Universitaria Careggi, Firenze, Italy

Introduction:

Quality of medical images is usually quantified by means

of some physical parameters. A different approach to assess the image quality

is the measurement of the performance of an observer on some specific

diagnostic task.

The purpose of this study is to show the potentiality of a method based

on Channelized Hotelling Observer (CHO) for the assessment of the per-

ceived image quality in order to optimize CT protocols.

Materials and Methods:

A method based on CHO model, reproducing the

response of a human observer, was used.

The image quality of abdominal oncology protocols in use on 6 different

CT scanners was evaluated and compared. The CHO method makes use of

a specific designed phantom, containing 5 cylindrical rods, with 5-mm di-

ameter and about 20 HU contrast at 120 kV, and of a mathematical algorithm

simulating the observer. The phantom was scanned on each scanner at 5

different dose levels around the value set on each protocol. The perfor-

mances of a scanner recently introduced in the clinical use were also

evaluated, depending on kernel filter and reconstruction method (FBP and

IR).

The CHO performances over the produced images were evaluated by lo-

calized ROC (L-ROC) analysis.

Results:

6 scanners from 3 different vendors were included. For each scanner,

a curve representing the scanner performances measured as AUC (Area

Under Curve) from L-ROC analysis versus CTDIvol was reported. The curves

were compared in order to highlight possible optimization requirements.

For the recently installed scanner useful indications about perceived image

quality in connection with different kernels and different levels of itera-

tive reconstruction were also obtained.

Conclusions:

Meaningful differences among scanners, especially among

scanners from different vendors, were found, suggesting the need for further

protocol optimization. From the curve relative to a scanner it was possible

to assess which dose level might be best suited for the given clinical

question.

http://dx.doi.org/10.1016/j.ejmp.2016.01.320

B.316

EVALUATION OF TECHNOLOGICAL INNOVATIONS IN STORAGE AND

DISTRIBUTION OF RADIOLOGICAL DATA

R. Tarducci

* , a ,

M. Pioppo

d ,

M. Duranti

b ,

P. Floridi

c ,

W. Orland

i e .

a

S.C. di Fisica

Sanitaria, Azienda Ospedaliera di Perugia, Perugia, Italy;

b

S.C. di Radiologia

1, Azienda Ospedaliera di Perugia, Perugia, Italy;

c

S.C. di Neuroradiologia,

Azienda Ospedaliera di Perugia, Perugia, Italy;

d

Direzione Sanitaria, Azienda

Ospedaliera di Perugia, Perugia, Italy;

e

Direzione Generale, Azienda Ospedaliera

di Perugia, Perugia, Italy

Introduction:

Santa Maria della Misericordia hospital is the largest hos-

pital in the region of Umbria. The hospital has more than 800 beds, and

42,000 patients were hospitalised in 2014.

Materials and Methods:

In 2014 the Azienda Ospedaliera di Perugia imple-

ments a solution that will make it possible to store data – images, reports,

etc – from the various hospital departments in the cloud, with a high level

of data privacy and security. At the same time they decided to implement

a patient portal, thus allowing patients access to their images and data, with

the same high levels of security. The aims was: (1) to increase the level of

data privacy and security; (2) to allow visualisation radiologic data using

numerous devices; (3) to reduce the number of patients returning to the

hospital to pick up their test results.

Results:

We have calculated the return on investment of cloud solution and

patients portal as financial savings in terms of spending on the: (1) storage

data and production of CDs and DVDs; and large systems of disaster re-

covery. We have calculated a 10% annual financial savings. Patients express

their appreciation of the fact that their data has been made more acces-

sible. The average web-access ratio over the first six months 2015 was 35.2%.

To understand the data we must take in account that in Italy the mean

number of families with internet access is 62.7%, whereas only 33.5% of

families use internet on a daily basis. Only 16.3% of elderly people have

internet access (ISTAT 2013).

Conclusion:

We would like to do the following:

Render the complete imaging archive accessible to patients; currently

only the latest exams are available for consulting, as part of transition

from a paper-based health record to an electronic health record;

Extend the service to the other hospitals throughout the region of

Umbria. This project has been the model to integrate data centres from

other hospital sites in the region;

Make instant messaging available to remind patients of future

appointments.

http://dx.doi.org/10.1016/j.ejmp.2016.01.321

B.317

DOSE MANAGEMENT PROGRAM: ONE YEAR RESULTS

G. Tosi

*

, M. Profili, A. Spinillo, L. Balzarini.

Istituto Clinico Humanitas, Rozzano

(MI), Italy

Purpose:

The goal of this study was to describe and analyze a project of

dose management from the beginning to the first results obtained.

Methods and Materials:

In late 2014 our radiology department in agree-

ment with the hospital’s leadership decided to implement a dose

management programwith the help of a software (Dosewatch™) and a train-

ing program.

A dose team was created, made by a radiologist, a physicist (team leader)

and a technician; the training program started at the beginning of 2015,

regarded initially the 2 CT equipment, and was based on four topic ele-

ments: standardization of protocols used, justification on dose levels that

exceeded a fixed dose threshold, optimization of acquisition protocols to

decrease doses and communication of obtained results at the hospital com-

munity and to the public/patients.

To actuate this program we defined an action plan in which it was clearly

evident what to do, who is responsible for, and a date dead line for each

step.

e93

Abstracts/Physica Medica 32 (2016) e71–e96