Atul Padole MD, Presenter: Nothing to Disclose

Efren Jesus Flores MD, Abstract Co-Author: Nothing to Disclose

Rachna Madan MD, Abstract Co-Author: Nothing to Disclose

Shelly Mishra, Abstract Co-Author: Nothing to Disclose

Sarabjeet Singh MD, Abstract Co-Author: Research Grant, Siemens AG Research Grant, Toshiba Corporation Research Grant, General Electric Company Research Grant, Koninklijke Philips NV

Sarvenaz Pourjabbar MD, Abstract Co-Author: Nothing to Disclose

Mannudeep K. S. Kalra MD, Abstract Co-Author: Nothing to Disclose

Ranish Deedar Ali Khawaja MD, Abstract Co-Author: Nothing to Disclose

Diego Alfonso Lira MD, Abstract Co-Author: Nothing to Disclose

Subba Rao Digumarthy MD, Abstract Co-Author: Nothing to Disclose

To assess image quality of chest CT reconstructed with image based (SafeCT), adaptive statistical (ASIR), & model based (MBIR) iterative reconstruction techniques (IRT) at less than 1 mGy CTDIvol.

Our IRB approved prospective study included 23 patients (mean age 63±13 years, 80±18 kg, M:F18:5) who underwent routine chest CT on a 64 channel MDCT (GE Discovery CT750 HD) and gave written informed consent for acquisition of ultra low dose (ULD) chest CT series. Standard chest CT (8±3.4 mGy) was followed by 3 ULD chest image series (0.2, 0.4, & 0.8 mGy) (total additional dose <1 mSv). Images were used to reconstruct SafeCT (CH0, CH1) and sinogram data were used to reconstructed with ASIR (SS70, SS90) & MBIR & standard CT with ASIR (SS40) (n=23*3*5+23=368 series). Board-certified thoracic radiologists performed independent & blinded evaluation for lesion detection, lesion conspicuity, & visibility of small structures from lowest to highest dose of ULD series & subsequently for standard dose CT.

Of 182 lesion, 112 non-calcified lung nodules (LN) & 8 ground glass opacities (GGO). There were 34 missed lesions [24 LN, 4GGO, 2 thyroid nodule (TN), 3 pleural effusions (PL)] at 0.2 mGy, 27 [18 LN, 2GGO, 2TN, 2 PL] at 0.4 mGy, & 11 [3LN,2GGO, 2TN, 2PL] at 0.8 mGy. The size of missed LN was less than 4mm. There were 7 & 4 false positive lesions at 0.2 & 0.4 mGy, respectively but none at 0.8 mGy. The conspicuity of LN was sufficient fo diagnostic performance for 3/19 at 0.2 mGy, 6/19 at 0.4 mGy & 10/17 (SafeCT:10,ASIR:10,MBIR:7) at 0.8 mGy. Visibility of sub-segmental bronchi was suboptimal at 0.2 & 0.4 mGy but sufficient for diagnostic performance at 0.8 mGy. Visibility of major fissure was suboptimal at 0.2&0.4 mGy but sufficient for 11/23 with IRT. Visibility of mediastinal and axillary lymph nodes was suboptimal at 0.2&0.4 mGy but sufficient for 9/23 with SafeCT, 8/23 with ASIR, 14/23 with MBIR at 0.8 mGy. Visibility of other mediastinal structures was limited at 0.8 mGy & suboptimal at 0.2&0.4 mGy.  

Most clinically significant lung lesions can be detected at CTDIvol of 0.8 mGy with SafeCT, ASIR, & MBIR. However, mediastinal structures could not be assessed with sufficient diagnostic confidence at 0.2-0.8 mGy with any IRT.

Lung nodules >4mm can be assessed with IRT at CTDIvol as low as 0.2 mGy but those < 4mm can be missed at CTDIvol less than 0.8 mGy regardless of the IRT.

Padole, A, Flores, E, Madan, R, Mishra, S, Singh, S, Pourjabbar, S, Kalra, M, Khawaja, R, Lira, D, Digumarthy, S, Assessment of Image Based, Adaptive Statistical, and Model Based Iterative Reconstruction Techniques for Chest CT at Less than 1 mGy CTDIvol.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14015147.html