NAME
m-ecat - CTI ECAT 6/7 medical image format (MedCon)
DESCRIPTION
This is a painful format. You should check the source code for more info.
There is only read support for ECAT 7. Below you will find the specs for the
ECAT 6 format. The ECAT 7 format differs a little in header definitions and
there is only one matrix entry per volume, while for ECAT 6 files there is
one matrix entry per plane. All ECAT 6 image data is written in one file
with `.img' extension.
-----------------------------------------------------------------------
Important Definitions
-----------------------------------------------------------------------
typedef struct mat_main_header {
char original_file_name[20];
Int16 sw_version;
Int16 data_type;
Int16 system_type;
Int16 file_type;
char node_id[10];
Int16 scan_start_day,
scan_start_month,
scan_start_year,
scan_start_hour,
scan_start_minute,
scan_start_second;
char isotope_code[8];
float isotope_halflife;
char radiopharmaceutical[32];
float gantry_tilt,
gantry_rotation,
bed_elevation;
Int16 rot_source_speed,
wobble_speed,
transm_source_type;
float axial_fov,
transaxial_fov;
Int16 transaxial_samp_mode,
coin_samp_mode,
axial_samp_mode;
float calibration_factor;
Int16 calibration_units,
compression_code;
char study_name[12],
patient_id[16],
patient_name[32],
patient_sex,
patient_age[10],
patient_height[10],
patient_weight[10],
patient_dexterity,
physician_name[32],
operator_name[32],
study_description[32];
Int16 acquisition_type,
bed_type,
septa_type;
char facility_name[20];
Int16 num_planes,
num_frames,
num_gates,
num_bed_pos;
float init_bed_position,
bed_offset[15],
plane_separation;
Int16 lwr_sctr_thres,
lwr_true_thres,
upr_true_thres;
float collimator;
char user_process_code[10];
Int16 acquisition_mode;
} Main_header;
#define MH_64_SIZE 446
typedef struct mat_scan_subheader {
Int16 data_type,
dimension_1,
dimension_2,
smoothing,
processing_code;
float sample_distance,
isotope_halflife;
Int16 frame_duration_sec;
Int32 gate_duration,
r_wave_offset;
float scale_factor;
Int16 scan_min,
scan_max;
Int32 prompts,
delayed,
multiples,
net_trues;
float cor_singles[16],
uncor_singles[16],
tot_avg_cor,
tot_avg_uncor;
Int32 total_coin_rate,
frame_start_time,
frame_duration;
float loss_correction_fctr;
Int32 phy_planes[8];
} Scan_subheader;
#define SSH_64_SIZE 236
typedef struct mat_image_subheader {
Int16 data_type,
num_dimensions,
dimension_1,
dimension_2;
float x_origin,
y_origin,
recon_scale, /* Image ZOOM from reconstruction */
quant_scale; /* Scale Factor */
Int16 image_min,
image_max;
float pixel_size,
slice_width;
Int32 frame_duration,
frame_start_time;
Int16 slice_location,
recon_start_hour,
recon_start_minute,
recon_start_sec;
Int32 gate_duration;
Int16 filter_code;
Int32 scan_matrix_num,
norm_matrix_num,
atten_cor_matrix_num;
float image_rotation,
plane_eff_corr_fctr,
decay_corr_fctr,
loss_corr_fctr,
intrinsic_tilt ;
Int16 processing_code,
quant_units,
recon_start_day,
recon_start_month,
recon_start_year;
float ecat_calibration_fctr,
well_counter_cal_fctr,
filter_params[6];
char annotation[40];
} Image_subheader;
#define ISH_64_SIZE 172
typedef struct mat_norm_subheader {
Int16 data_type,
dimension_1,
dimension_2;
float scale_factor;
Int16 norm_hour,
norm_minute,
norm_second,
norm_day,
norm_month,
norm_year;
float fov_source_width;
float ecat_calib_factor;
} Norm_subheader;
#define NSH_64_SIZE 30
typedef struct mat_attn_subheader {
Int16 data_type,
attenuation_type,
dimension_1,
dimension_2;
float scale_factor,
x_origin,
y_origin,
x_radius,
y_radius,
tilt_angle,
attenuation_coeff,
sample_distance;
} Attn_subheader;
#define ASH_64_SIZE 40
-----------------------------------------------------------------------
What does the format support or not support:
===========================================================================
Item Supported Not Supported
===========================================================================
Color Map : grayscale -
File Endian : big little
Pixeltypes : VAX Int16 (write) | All (read) -
===========================================================================
Scaling factors : quantify & calibrate factors/image are supported
---------------------------------------------------------------------------
Dimensions/Image : different dimensions for each image are NOT supported
---------------------------------------------------------------------------
Pixeltypes/Image : different pixeltypes for each image are NOT supported
===========================================================================
NOTES
The MedCon program also supports the reading of sinogram, attenuation and
normalization files for conversion purposes but it does not support writing
those file types. In fact, they will be considered as reconstructed data!
The format supports more pixeltypes. The reason for our restriction of
writing only the Int16 type was our ECAT software that only supports the
Int16 type.
We consider three kinds of images (planes) in an ECAT file:
(1) plain pixel values [no unit] (ppv = ppv)
- the planes/images are NOT normalized
(2) quantified values [counts/second/pixel] (qpv = ppv * quant_scale)
- the planes/images are normalized
(3) calibrated values [uCi/ml] (cpv = qpv * calibr_fctr)
- the planes/images are normalized
The float values in the header are stored as VAX format.
FILES
/usr/local/xmedcon/source/m-ecat64.h The header file.
/usr/local/xmedcon/source/m-ecat64.c The source file.
/usr/local/xmedcon/source/m-matrix64.h CTI header file.
/usr/local/xmedcon/source/m-matrix64.c CTI source file.
SEE ALSO
medcon(1), xmedcon(1), xmedcon-config(1)
m-acr(4), m-anlz(4), m-gif(4), m-inw(4), m-intf(4)
medcon(3)
AUTHOR
(X)MedCon project was originally written by Erik Nolf (eNlf) for the former
PET-Centre at Ghent University (Belgium).
e-mail: enlf-at-users.sourceforge.net www: http://xmedcon.sourceforge.net