C a r d i o p u l m o n a r y I m a g i n g • Te c h n i c a l I n n ov a t i o n

Hochhegger et al.
Chemical-Shift MRI of Pulmonary Hamartomas

Cardiopulmonary Imaging
Technical Innovation

Chemical-Shift MRI of Pulmonary

Downloaded from ajronline.org by 180.252.175.185 on 02/15/24 from IP address 180.252.175.185. Copyright ARRS. For personal use only; all rights reserved

Hamartomas: Initial Experience

Using a Modified Technique to
Assess Nodule Fat
Bruno Hochhegger 1,2 OBJECTIVE. The aim of this study was to show the usefulness of chemical-shift MRI
Edson Marchiori2 in the diagnosis of intranodular fat in seven patients with pulmonary hamartomas and inde-
Daniela Quinto dos Reis1 terminate CT findings.
Arthur Soares Souza, Jr. 3 CONCLUSION. In the setting of chemical-shift MRI, the average nodule signal inten-
Luciana Soares Souza3 sity index of pulmonary hamartomas was 45.3% (SD = 25.5%). The correlation between av-
erage nodule signal intensity and CT attenuation in Hounsfield units was –0.94. Chemical-
Taua Brum1
shift MRI could be an important tool for the detection of fat in pulmonary hamartomas with
Klaus L. Irion 4 inconclusive CT findings.
Hochhegger B, Marchiori E, Quinto dos Reis D,
et al.

Keywords: chemical-shift MRI, CT, diagnostic radiology,
lung neoplasm, MRI, pulmonary hamartoma
DOI:10.2214/AJR.11.8056
Received October 10, 2011; accepted after revision
January 10, 2012.
1
Chest Radiology Department, Santa Casa de Porto
Alegre, Rua 24 de Outubro 925/903, Bairro Moinhos de
Vento, Porto Alegre, Rio Grande do Sul, 90510-002,
Brazil. Address correspondence to B. Hochhegger
(brunohochhegger@gmail.com).
2
Department of Radiology, Universidade Federal do Rio
de Janeiro, Rio de Janeiro, Brazil.
3
São Jose do Rio Preto School of Medicine, Porto Alegre,
Rio Grande do Sul, Brazil.
4
Liverpool Heart and Chest Hospital NHS Foundation
Trust, Liverpool, England.
WEB
This is a Web exclusive article.
AJR 2012; 199:W331–W334
0361–803X/12/1993–W331
© American Roentgen Ray Society
P
ulmonary hamartomas are the
most common benign tumors of
the lung and the third most fre-
quent cause of solitary pulmo-
nary nodules, accounting for approximately
8% of primary lung tumors [1–3]. The radio-
logic diagnosis of pulmonary hamartomas is
frequently made using CT, especially CT
densitometry. This technique is one of the
most sensitive methods of eliminating partial
volume averaging for the detection of intra-
nodular fat, thus providing a highly predic-
tive diagnosis of a pulmonary hamartoma [1,
2]. Nevertheless, when CT shows neither fat
nor calcifications, the diagnosis of pulmo-
nary hamartomas without invasive means is
very difficult [1].
MRI of the lung has shown important
progress in speed and image quality in recent
years and is now appropriate for routine clin-
ical use. MRI with the chemical-shift tech-
nique can identify intracellular lipids due to
the different resonant frequencies of fat and
water protons in a given voxel [4]. However,
to our knowledge, no report has assessed the
use of chemical-shift MRI for the diagnosis
of pulmonary hamartomas.
The aim of this study was to show the use-
fulness of chemical-shift MRI for the diag-
nosis of intranodular fat in pulmonary ham-
artomas and to compare the results of this
imaging technique with CT findings.
Materials and Methods
Patients
After receiving institutional research ethics
board approval, we identified all patients who had
undergone thoracic CT for the characterization of
pulmonary nodules between January 1 and Au-
gust 15, 2011, and who had CT findings showing
negative densities. All thoracic CT examinations
during this period (n = 1300) were evaluated by
two radiologists, one with 15 years of experience
and the other with 8 years; they performed den-
sity measurements of all nodules larger than 0.7
cm. From review of the 1300 CT examinations, 15
nodules were found.
Eight of the 15 selected patients were exclud-
ed from this study because the CT reports con-
cluded that the lowest nodule attenuation was less
than –40 HU in at least 8 pixels, which is consid-
ered consistent with a hamartoma in our clinical
practice and does not necessitate further imaging
[1]. One thoracic radiologist with 8 years of ex-
perience reviewed the CT and MR images of the
other seven patients who had undergone chemical-
shift MRI. All seven nodules that had indetermi-
nate CT findings were surgically resected and had
pathologic diagnoses of hamartoma.
CT
Unenhanced CT was performed on a 64-MDCT
scanner (LightSpeed VCT, GE Healthcare). The CT
parameters included a slice collimation of 0.625 mm,
interval reconstruction of 0.625 mm, high-quality

AJR:199, September 2012 W331

 Hochhegger et al.

mode, table speed of 7.5 mm/s, 120 kV, and 200
mA. The density measurement was performed
with a region-of-interest (ROI) tool, and the low-
est density of a nodule was defined as the lowest
density that extended for or covered at least 8 con-
tiguous pixels.
Downloaded from ajronline.org by 180.252.175.185 on 02/15/24 from IP address 180.252.175.185. Copyright ARRS. For personal use only; all rights reserved
imaging, and diffusion-weighted imaging (DWI)
sequences. T1-weighted fast spin-echo (TR/TE,
480/10) and T2-weighted fast spin-echo (3200/85)
sequences were performed, both axial fat-sup-
pressed, with a 35-cm FOV, 5-mm slice thickness
with a 2.5-mm gap, and a 128–192 × 128–224 ma-
trix. DWI was performed in the axial plane using a
slice thickness, 7 mm; 1-mm interslice gap; and
2–3 excitations. In- and opposed-phase MR im-
ages were also obtained in the transverse plane in
one breath-hold with a 2D spoiled gradient-echo
sequence with dual-echo acquisition using the fol-
lowing parameters: TR range/first-echo TE range,
second-echo TE range, 80–200/4.2–4.6, 2.1–2.3;

MRI breath-hold single-shot spin-echo echo-planar im- bandwidth, 31.25–62.5 kHz; 1 signal acquired;
MRI was performed on a 1.5-T scanner (HDxT aging acquisition. DWI parameters included a b section thickness, 5 mm; gap, 0 mm; FOV to cov-
Excite, GE Healthcare). The basic MR sequences value of 50 and 600 s/mm2; TR range/TE range, er the nodules, 24–38 cm; matrix, 256 × 160–192;
were transverse T1-weighted imaging, T2-weighted 2800–3500/50–58; matrix, 128–192 × 128–224; and flip angle, 75–90°.

A B C

D E F
Fig. 1—60-year-old man with colon cancer.
A, Sagittal volume-rendering CT reconstruction shows nodule.
B, Axial CT image shows low-attenuation areas in nodule. This nodule had low attenuation of –33 HU.
C, T1-weighted image shows internal septations.
D and E, T2-weighted (D) and in-phase (E) images show high signal intensity of nodule.
F, Opposed-phase image shows loss of signal in nodule.

W332 AJR:199, September 2012

 Chemical-Shift MRI of Pulmonary Hamartomas
One observer with 8 years of experience in
MRI interpretation performed all quantitative
MRI measurements. This observer was unaware
of the CT attenuation measurements. ROIs were
placed on each nodule using the largest ellipse
possible while avoiding the edges—where T2*
Downloaded from ajronline.org by 180.252.175.185 on 02/15/24 from IP address 180.252.175.185. Copyright ARRS. For personal use only; all rights reserved
artifacts could be present. Identical ROIs were
drawn on in- and opposed-phase MR images us-
ing the copy-and-paste functions of the worksta-
tion. A similar method was used to measure the
signal intensity of the longissimus dorsi muscle
while avoiding any artifact or subcutaneous fat.
The muscle was used instead of the spleen be-
cause one of our patients had undergone a sple-
nectomy and another had hemosiderosis.
The percentage of signal intensity decrease was
calculated using the nodule signal intensity index
(nodule lipid index):
[(SIIP − SIOP) / SIIP] × 100,
where SIIP and SIOP are the signal intensities of the
nodule measured on in- and opposed-phase imag-
es, respectively [5]. An index value of more than
17% was considered consistent with a lipid-rich
lesion in a study of adrenal lesions [5]; we also
used 17% as a cutoff in our study.
Microsoft Excel 2000 software was used for
data collection and to calculate correlations. All
statistical analyses were performed using SPSS
software (version 13.0, SPSS) for Microsoft Win-
dows, with p < 0.05 considered to indicate a sig-
nificant difference.
Results
All seven cases (five men and two wom-
en; age range, 48–65 years) were diagnosed
pathologically as hamartomas. No case had
characteristic calcifications, and all lesions
showed internal fat on histopathologic analy-
sis. The average diameter of the lesions was
1.52 cm (SD = 0.56 cm). All lesions had lob-
ulated margins and showed no pulmonary
parenchymal infiltration.
On CT (Figs. 1A and 1B), no case showed
calcium attenuation and the average lowest
density was –21.4 HU (SD = 9.6 HU) (Table 1).
All cases showed intermediate signaling
(i.e., lower than fat and higher than spinal
muscle) on T1-weighted MRI sequences. On
T2-weighted sequences, all lesions showed
high intensity signals and three had internal
septations. The ratio of lesions to paraspinal
muscle exceeded 3 in all cases. In the set-
ting of chemical-shift MRI (Figs. 1C–1F),
the average nodule signal intensity index was
45.3% (SD = 25.5%) (Table 1). The correla-
tion between average nodule signal intensity
and CT attenuation was –0.94. This correla-
AJR:199, September 2012
tion is good; it shows that the nodule signal
intensity index is associated with negative
densities on CT. No nodule showed restric-
tion in DWI sequences.
Discussion
The diagnosis of a pulmonary hamartoma
using CT is heavily dependent on the pres-
ence of detectable fat within the lesion. How-
ever, investigators have estimated that only
approximately 60% of hamartomas contain
significant fat deposits that are visible on CT
[1]. Characteristic CT findings of hamarto-
mas are a lesion that is 2.5 cm in diameter
or smaller, has a smooth edge, and has fo-
cal areas of fat (–40 to –120 HU in at least 8
voxels) or areas of calcification (> 175 HU)
alternating with areas of fat in the lesion [1].
MRI can identify intracellular lipid be-
cause the resonant frequencies of fat and wa-
ter protons in a given voxel are different [4,
6]. Because fat protons precess at a lower
frequency than do water protons, they cancel
each other out during out-of-phase breath-
hold gradient-echo MRI [5, 7]. This phe-
nomenon results in loss of signal intensity on
out-of-phase imaging in comparison with in-
phase imaging. Previous reports have shown
an inverse linear relationship between the
percentage of lipid-rich cells and the relative
change in signal intensity on chemical-shift
MRI [4]. Chemical-shift MRI has become a
popular technique for the diagnosis of adre-
nal adenomas [4].
Fat-water separation is an important capa-
bility of chemical-shift imaging. A general
spectroscopic imaging approach is certainly
possible but can become very time-consum-
ing for the desired spatial coverage and reso-
lution. Many methods that take advantage of
the prior information were developed over the
past 2 decades to deal exclusively with fat and
TABLE 1: CT and MRI Characteristics of Pulmonary Nodules Confirmed
to Be Hamartomas at Pathology
Nodule
Case Lowest Lipid
No. Diameter (cm) Attenuation (HU) Index (%)
1 2.5 −8 26
2 1.5 −33 75
3 0.9 −16 27
4 1.3 −25 40
5 0.8 −35 80
6 2.1 −11 24
7 1.6 −22 35
water imaging, starting with Dixon’s original
proposal of a simple two-point method [8].
To improve the robustness of Dixon’s meth-
od and to address field inhomogeneity, some
investigators have proposed multipoint fat-
water separation techniques that are quite ef-
fective at separating fat and water signals but
that double or triple scanning time. Therefore,
signal-to-noise ratio–efficient acquisition and
processing are required to warrant the extra
scanning time [9]. There has also been inter-
est in using multiecho sequences to reduce
the total scanning time required for acquiring
multiple time points [10]. Multiecho acquisi-
tions can also improve spatial registration of
the acquired images.
Almost 60% of pulmonary hamartomas
contain fat, and this structural characteristic
has been used as a safe criterion of benignity.
In previous studies, investigators have used
chemical-shift imaging to study pulmonary
nodules [11], although, to our knowledge,
this study is the first to use chemical-shift
MRI in the detection of fat in pulmonary
hamartomas. Our data suggest that chemi-
cal-shift MRI could be an important tool to
use to detect fat in pulmonary hamartomas
with inconclusive CT findings.
Our study also indicated that hamartomas
show intermediate signaling on T1-weight-
ed sequences, high signal intensity on T2-
weighted sequences, and no signal restriction
on DWI sequences. Three of the seven nod-
ules contained septations with signal inten-
sities lower than those of the main tumor on
T2-weighted sequences. All of these findings
have been reported previously [12].
Our study has some limitations. First, the
number of cases was small and further analy-
sis is necessary to confirm the data. Second, we
studied only hamartomas with inconclusive CT
findings, so the findings are not applicable to
Ratio of Lesion–Paraspinal Muscle
Signal on Diffusion-
T2 Signal T1 Signal Weighted Imaging
5.2 1.0 No restriction
4.1 0.5 No restriction
3.9 0.9 No restriction
3.1 1.0 No restriction
3.5 0.5 No restriction
4.5 0.9 No restriction
3.5 0.8 No restriction
W333

all hamartomas. Finally, our series contained no
case of a nodule smaller than 0.8 cm in diam-
eter; chemical-shift MRI has been reported
to have a lower accuracy in the detection of
small lesions [7].
In conclusion, chemical-shift MRI could
be an important tool to detect fat in pulmo-
Downloaded from ajronline.org by 180.252.175.185 on 02/15/24 from IP address 180.252.175.185. Copyright ARRS. For personal use only; all rights reserved
nary hamartomas with inconclusive CT find-
ings. Our data also confirmed that pulmonary
hamartomas had high signal intensity on T2-
weighted sequences and showed no restriction
in signal intensity on DWI sequences.
References
1. Siegelman SS, Khouri NF, Scott WW Jr, et al.
Pulmonary hamartoma: CT findings. Radiology
1986; 160:313–317
2. Bateson EM. So-called hamartoma of the lung: a
W334
Hochhegger et al.
true neoplasm of fibrous connective tissue of the
bronchi. Cancer 1973; 31:1458–1467
3. Hochhegger B, Marchiori E, Sedlaczek O, et al.
MRI in lung cancer: a pictorial essay. Br J Radiol
2011; 84:661–668
4. Blake MA, Cronin CG, Boland GW. Adrenal im-
aging. AJR 2010; 194:1450–1460
5. Fujiyoshi F, Nakajo M, Kukukura Y, Tsuchimochi
S. Characterization of adrenal tumors by chemi-
cal shift fast low-angle shot MR imaging: com-
parison of four methods of quantitative evalua-
tion. AJR 2003; 180:1649–1657
6. Outwater EK, Siegelman ES, Radecki PD, Piccoli
CW, Mitchell DG. Distinction between benign and
malignant adrenal masses: value of T1-weighted
chemical shift MR imaging. AJR 1995; 165:579–583
7. Israel GM, Korobkin M, Wang C, Hecht EN,
Krinsky GA. Comparison of unenhanced CT and
chemical shift MRI in evaluating lipid-rich adre-
nal adenomas. AJR 2004; 183:215–219
8. Dixon WT. Simple proton spectroscopic imaging.
Radiology 1984; 153:189–194
9. Glover GH, Schneider E. Three-point Dixon technique
for true water/fat decomposition with B0 inhomogene-
ity correction. Magn Reson Med 1991; 18:371–383
10. Ma J. Breath-hold water and fat imaging using
dual-echo two-point Dixon technique with an ef-
ficient and robust phase-correction algorithm.
Magn Reson Med 2004; 52:415–419
11. Satoh S, Kitazume Y, Ohdama S, Kimula Y, Tau-
ra S, Endo Y. Can malignant and benign pulmo-
nary nodules be differentiated with diffusion-
weighted MRI? AJR 2008; 191:464–470
12. Sakai F, Sone S, Kiyono K, et al. MR of pulmo-
nary hamartoma: pathologic correlation. J Thorac
Imaging 1994; 9:51–55
AJR:199, September 2012
 This article has been cited by:

1. Moritz T. Winkelmann, Sebastian Gassenmaier, Sven S. Walter, Christoph Artzner, Konstantin Nikolaou, Malte N. Bongers.
2024. Differentiation of Hamartomas and Malignant Lung Tumors in Single-Phased Dual-Energy Computed Tomography.
Tomography 10:2, 255-265. [Crossref]
2. Yang Li, Li Yang, Xiaolong Gu, Yaning Wang, Huiyan Deng, Hui Feng, Ning Zhang, Mingbo Wang, Qi Wang, Gaofeng
Shi. 2023. Magnetic resonance imaging findings of pulmonary sclerosing pneumocytoma: a case report and literature review.
Downloaded from ajronline.org by 180.252.175.185 on 02/15/24 from IP address 180.252.175.185. Copyright ARRS. For personal use only; all rights reserved

Frontiers in Oncology 13. . [Crossref]

3. Alessandra Borgheresi, Andrea Agostini, Luca Pierpaoli, Alessandra Bruno, Tommaso Valeri, Ginevra Danti, Eleonora Bicci,
Michela Gabelloni, Federica De Muzio, Maria Chiara Brunese, Federico Bruno, Pierpaolo Palumbo, Roberta Fusco, Vincenza
Granata, Nicoletta Gandolfo, Vittorio Miele, Antonio Barile, Andrea Giovagnoni. 2023. Tips and Tricks in Thoracic Radiology
for Beginners: A Findings-Based Approach. Tomography 9:3, 1153-1186. [Crossref]
4. Bruno Hochhegger, Matheus Zanon, Pratik P Patel, Nupur Verma, Diego André Eifer, Pedro Paulo Teixeira e Silva Torres,
Arthur S Souza, Luciana Volpon Soares Souza, Tan-Lucien Mohammed, Edson Marchiori, Jeanne B Ackman. 2022. The
diagnostic value of magnetic resonance imaging compared to computed tomography in the evaluation of fat-containing thoracic
lesions. The British Journal of Radiology 95:1140. . [Crossref]
5. Jeanne B. Ackman. 2022. Beyond Chest Computed Tomography. Advances in Clinical Radiology 4:1, 73-85. [Crossref]
6. Emre Ünal, Sevtap Arslan, Gulnar Aghayeva, Yasin Sarıkaya, Türkmen Çifti, Sevgen Önder, Devrim Akıncı, Okan Akhan.
2021. Rare Pulmonary Tumors and Carcinoma Mimickers; Experience from an Interventional Radiology Unit with Radiologic-
pathologic Correlation- A Pictoral Essay. Current Medical Imaging Formerly Current Medical Imaging Reviews 17:10, 1183-1190.
[Crossref]
7. Koji Takumi, Hiroaki Nagano, Tomohiro Harasawa, Kazuhiro Tabata, Takuya Tokunaga, Takashi Yoshiura. 2021. Pulmonary
hamartoma: Feasibility of dual-energy CT detection of intranodular fat. Radiology Case Reports 16:5, 1032-1036. [Crossref]
8. Eric S. Bih, Monica Epelman, Ricardo Restrepo, Edward Y. Lee. Lung 173-193. [Crossref]
9. Mark C. Liszewski, Pierluigi Ciet, Edward Y. Lee. Lung and Pleura 1-28. [Crossref]
10. Anup S. Shetty, Adam L. Sipe, Maria Zulfiqar, Richard Tsai, Demetrios A. Raptis, Constantine A. Raptis, Sanjeev Bhalla.
2019. In-Phase and Opposed-Phase Imaging: Applications of Chemical Shift and Magnetic Susceptibility in the Chest and
Abdomen. RadioGraphics 39:1, 115-135. [Crossref]
11. Jan Pojda, Sandra Hürlimann, Alfred Leiser. 2019. Giant pulmonary chondroid hamartoma in a 74-year-old female patient. A
case report and brief review of literature. Journal of Lung, Pulmonary & Respiratory Research 6:3, 50-53. [Crossref]
12. Bruno Hochhegger, Matheus Zanon, Stephan Altmayer, Gabriel S. Pacini, Fernanda Balbinot, Martina Z. Francisco, Ruhana
Dalla Costa, Guilherme Watte, Marcel Koenigkam Santos, Marcelo C. Barros, Diana Penha, Klaus Irion, Edson Marchiori.
2018. Advances in Imaging and Automated Quantification of Malignant Pulmonary Diseases: A State-of-the-Art Review. Lung
196:6, 633-642. [Crossref]
13. Yuji Kishida, Hisanobu Koyama, Shinichiro Seki, Takeshi Yoshikawa, Katsusuke Kyotani, Tomoyuki Okuaki, Kazuro Sugimura,
Yoshiharu Ohno. 2018. Comparison of fat suppression capability for chest MR imaging with Dixon, SPAIR and STIR
techniques at 3 Tesla MR system. Magnetic Resonance Imaging 47, 89-96. [Crossref]
14. Annemie Snoeckx, Pieter Reyntiens, Damien Desbuquoit, Maarten J. Spinhoven, Paul E. Van Schil, Jan P. van Meerbeeck,
Paul M. Parizel. 2018. Evaluation of the solitary pulmonary nodule: size matters, but do not ignore the power of morphology.
Insights into Imaging 9:1, 73-86. [Crossref]
15. Juergen Biederer. General Requirements of MRI of the Lung and Suggested Standard Protocol 1-20. [Crossref]
16. Takuya Tokunaga, Naoya Yokomakura, Go Kamimura, Koji Takumi, Yoshihiro Nakamura, Masami Sato. 2017. A case of
enucleated pulmonary hamartoma -Potential usefulness of chemical shift MRI-. The Journal of the Japanese Association for Chest
Surgery 31:4, 506-510. [Crossref]
17. Miriam Menna-Barreto, Gláucia Zanetti, Edson Marchiori. 2016. The Role of Imaging Methods in the Diagnosis of Pulmonary
Lipoma. Archivos de Bronconeumología (English Edition) 52:4, 223. [Crossref]
18. Miriam Menna-Barreto, Gláucia Zanetti, Edson Marchiori. 2016. Papel de las pruebas de imagen en el diagnóstico del lipoma
pulmonar. Archivos de Bronconeumología 52:4, 223. [Crossref]
19. Bruno Hochhegger, Carlos S. Nin, Giordano R.T. Alves, Daniela R. Hochhegger, Vinícius V.S. de Souza, Guilherme Watte,
Klaus L. Irion, Marcos D. Guimarães, Edson Marchiori. 2016. Multidetector Computed Tomography Findings in Pulmonary
Hamartomas. Journal of Thoracic Imaging 31:1, 11-14. [Crossref]
 20. Bruno Hochhegger, Vinícius Valério Silveira de Souza, Edson Marchiori, Klaus Loureiro Irion, Arthur Soares Souza Jr., Jorge
Elias Junior, Rosana Souza Rodrigues, Miriam Menna Barreto, Dante Luiz Escuissato, Alexandre Dias Mançano, César Augusto
Araujo Neto, Marcos Duarte Guimarães, Carlos Schuler Nin, Marcel Koenigkam Santos, Jorge Luiz Pereira e Silva. 2015. Chest
magnetic resonance imaging: a protocol suggestion. Radiologia Brasileira 48:6, 373-380. [Crossref]
21. Carolina A. Souza. 2015. MRI of the chest: review of imaging strategies. Radiologia Brasileira 48:6, V-VI. [Crossref]
22. Brian J. Bartholmai, Chi Wan Koo, Geoffrey B. Johnson, Darin B. White, Sushravya M. Raghunath, Srinivasan Rajagopalan,
Michael R. Moynagh, Rebecca M. Lindell, Thomas E. Hartman. 2015. Pulmonary Nodule Characterization, Including
Downloaded from ajronline.org by 180.252.175.185 on 02/15/24 from IP address 180.252.175.185. Copyright ARRS. For personal use only; all rights reserved

Computer Analysis and Quantitative Features. Journal of Thoracic Imaging 30:2, 139-156. [Crossref]
23. Yasuyuki Kurihara, Shin Matsuoka, Tsuneo Yamashiro, Atsuko Fujikawa, Shoichiro Matsushita, Kunihiro Yagihashi, Yasuo
Nakajima. 2014. MRI of Pulmonary Nodules. American Journal of Roentgenology 202:3, W210-W216. [Abstract] [Full Text]
[PDF] [PDF Plus]
24. Wanglae Cho, Yeon Joo Jeong, Chang Hun Lee, Ji Won Lee, Kun-Il Kim, Jeong A Yeom, Yeong Dae Kim. 2013. Dynamic CT
Findings of Pulmonary Hamartoma: A Comparison with Histopathologic Findings. Journal of the Korean Society of Radiology
69:6, 449. [Crossref]
25. Ayako Hamano, Kazuhiro Suzuki, Tsuyoshi Saito, Ryohei Kuwatsuru, Shiaki Oh, Kenji Suzuki. 2013. Infiltrating angiolipoma
of the thoracic wall: A case report. Open Journal of Clinical Diagnostics 03:02, 19-22. [Crossref]