Liver segmentation is relevant for several clinical applications. Automatic liver segmentation using convolutional neural networks (CNNs) has been recently investigated. In this paper, we propose a new approach of combining a largest connected component (LCC) algorithm, as a post-processing step, with CNN approaches to improve liver segmentation accuracy. Specifically, in this study, the algorithm is combined with three well-known CNNs for liver segmentation: FCN-CRF, DRIU and V-net. We perform the experiment on a variety of liver CT images, ranging from non-contrast enhanced CT images to low-dose contrast enhanced CT images. The methods are evaluated using Dice score, Haudorff distance, mean surface distance, and false positive rate between the liver segmentation and the ground truth. The quantitative results demonstrate that the LCC algorithm statistically significantly improves results of the liver segmentation on non-contrast enhanced and low-dose images for all three CNNs. The combination with V-net shows the best performance in Dice score (higher than 90%), while the DRIU network achieves the smallest computation time (2 to 6 seconds) for a single segmentation on average. The source code of this study is publicly available at https://github.com/kennyha85/Liver-segmentation. Keywords: Liver segmentations, CNNs, Connected Components, Post processing Reference [1] K.A. McGlynn, J.L. Petrick, W.T. London, Global epidemiology of hepatocellular carcinoma: An emphasis on demographic and regional variability. Clinics in liver disease 19(2) (2015) 223-238. [2] M. Mohammadian, N. Mahdavifar, A. Mohammadian-Hafshejani, H. Salehiniya, Liver cancer in the world: epidemiology, incidence, mortality and risk factors, World Cancer Res J. 5(2) (2018) e1082. [3] T.T. Hong, N. Phuong Hoa, S.M. Walker, P.S. Hill, C. Rao, Completeness and reliability of mortality data in Viet Nam: Implications for the national routine health management information system, PloS one 13(1) 2018) e0190755. https://doi.org/10.1371/journal.pone.0190755. [4] T. Pham, L. Bui, G. Kim, D. Hoang, T. Tran, M. Hoang, Cancers in Vietnam-Burden and Control Efforts: A Narrative Scoping Review. Cancer Control 26(1) (2019) 1073274819863802. [5] M. Borner, M. Castiglione, J. Triller, H.U. Baer, M. Soucek, L. Blumgart, K. Brunner, Arena: Considerable side effects of chemoembolization for colorectal carcinoma metastatic to the liver, Annals of oncology 3(2) (1992) 113-115. [6] K. Memon, R.J. Lewandowski, L. Kulik, A. Riaz, M.F. Mulcahy, R. Salem, Radioembolization for primary and metastatic liver cancer, In Seminars in radiation oncology, WB Saunders. 21(4) (2011) 294-302. [7] I. Gory, M. Fink, S. Bell, P. Gow, A. Nicoll, V. Knight, W. Kemp, Radiofrequency ablation versus resection for the treatment of early stage hepatocellular carcinoma: A multicenter Australian study, Scandinavian journal of gastroenterology 50(5) (2015) 567-576. [8] H.M. Luu, C. Klink, W. Niessen, A. Moelker, T. Van Walsum, Non-rigid registration of liver CT images for CT-guided ablation of liver tumors. PloS one, 11(9) 92016) e0161600. [9] G. Gunay, M.H. Luu, A. Moelker, T. Van Walsum, S. Klein, Semiautomated registration of pre‐and intraoperative CT for image‐guided percutaneous liver tumor ablation interventions, Medical physics 44(7) (2017) 3718-3725. [10] A. Gotra, L. Sivakumaran, G. Chartrand, N. Vu, F. Vandenbroucke-Menu, C. Kauffmann, A. Tang, Liver segmentation: Indications, techniques and future directions, Insights into imaging 8(4) (2017) 377-392. https://doi.org/10.1007/s13244-017-0558-1. [11] T. Heimann, B. Van Ginneken, M.A. Styner, Y. Arzhaeva, V. Aurich, C. Bauer, F. Bello, Comparison and evaluation of methods for liver segmentation from CT datasets, IEEE transactions on medical imaging 28(8) (2009) 1251-1265. [12] O. Ronneberger, P. Fischer, T. Brox, U-net: Convolutional networks for biomedical image segmentation, In International Conference on Medical image computing and computer-assisted intervention, Springer, Cham, 2015, pp. 234-241. [13] F. Milletari, N. Navab, S.A. Ahmadi, October, V-net: Fully convolutional neural networks for volumetric medical image segmentation, In 2016 Fourth International Conference on 3D Vision (3DV) IEEE, 2016, pp. 565-571. [14] P.F. Christ, F. Ettlinger, F. Grün, M.A. Elshaera, J. Lipkova, S. Schlecht, M. Rempfler, Automatic liver and tumor segmentation of CT and MRI volumes using cascaded fully convolutional neural networks, arXiv preprint arXiv:1702.05970, 2017. [15] P.F. Christ, M.E.A. Elshaer, F. Ettlinger, S. Tatavarty, M. Bickel, P. Bilic, H. Sommer, Automatic liver and lesion segmentation in CT using cascaded fully convolutional neural networks and 3D conditional random fields. In International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer, Cham, 2016, pp. 415-423. [16] H. Meine, G. Chlebus, M. Ghafoorian, I. Endo, A. Schenk, Comparison of U-net-based Convolutional Neural Networks for Liver Segmentation in CT. arXiv preprint arXiv, 2018, pp. 1810.04017. [17] X. Li, H. Chen, X. Qi, Q. Dou, C.W. Fu, P.A. Heng, H-DenseUNet: hybrid densely connected UNet for liver and tumor segmentation from CT volumes, IEEE transactions on medical imaging, 37(12) (2018) 2663-2674. [18] M. Bellver, K.K. Maninis, J. Pont-Tuset, X. Giró-i-Nieto, J. Torres, L. Van Gool, Detection-aided liver lesion segmentation using deep learning, ArXiv preprint arXiv:1711.11069, 2017. [19] H.S. Hoang, C.P. Pham, D. Franklin, T. Van Walsum, M.H. Luu, An Evaluation of CNN-based Liver Segmentation Methods using Multi-types of CT Abdominal Images from Multiple Medical Centers, In 2019 19th International Symposium on Communications and Information Technologies (ISCIT), IEEE, September, 2019, pp. 20-25. [20] H. Samet, M. Tamminen, Efficient component labeling of images of arbitrary dimension represented by linear bintrees, IEEE Transactions on Pattern Analysis and Machine Intelligence, 10(4) (1988) 579-586. [21] P. Bilic, P.F. Christ, E. Vorontsov, G. Chlebus, H. Chen, Q. Dou, S. Kadoury, The liver tumor segmentation benchmark (lits), ArXiv preprint arXiv, 2019, 1901.04056. [22] H.M. Luu, A. Moelker, S. Klein, W. Niessen, T. Van Walsum, Quantification of nonrigid liver deformation in radiofrequency ablation interventions using image registration, Physics in Medicine & Biology 63(17) (2018) 175005. [23] A.A. Novikov, D. Major, M. Wimmer, D. Lenis, K. Bühler, Deep Sequential Segmentation of Organs in Volumetric Medical Scans, IEEE transactions on medical imaging, 2018. [24] Y. Huo, J.G. Terry, J. Wang, S. Nair, A. Lasko, B.I. Freedman, B.A. Landman, Fully Automatic Liver Attenuation Estimation combing CNN Segmentation and Morphological Operations, Medical physics, 2019. [25] N. Gruber, S. Antholzer, W. Jaschke, C. Kremser, M. Haltmeier, A Joint Deep Learning Approach for Automated Liver and Tumor Segmentation, ArXiv preprint arXiv, 2019, pp. 1902.07971. [26] S. Chen, K. Ma, Y. Zheng, Med3D: Transfer Learning for 3D Medical Image Analysis, ArXiv preprint arXiv, 2019, pp. 1904.00625. [27] W. Tang, D. Zou, S. Yang, J. Shi, DSL: Automatic Liver Segmentation with Faster R-CNN and DeepLab, In International Conference on Artificial Neural Networks, Springer, Cham, 2018, pp. 137-147.

No abstract

Background Heroin use continues to drive HIV transmission in Vietnam, but methamphetamine and alcohol use are growing rapidly and, as in other countries, polysubstance use is widespread. The objective of this study was to understand the interplay between heroin, methamphetamine, and alcohol use among people with opioid use disorder (OUD) and HIV in Vietnam. Methods We conducted 44 in-depth, face-to-face qualitative interviews with people with OUD and HIV who participated in the BRAVO trial of buprenorphine versus methadone in five Vietnam HIV clinics. Interviews probed participants’ experiences of heroin, methamphetamine, and alcohol use and their interplay with HIV/OUD treatment. Interviews were professionally transcribed and analyzed using a thematic analysis approach. Results Of 44 participants interviewed 42 were male, on average 38.8 years of age, with 30 reporting a history of methamphetamine use and 33 reporting a history of alcohol use. Several themes emerged: 1) Methamphetamine and alcohol were perceived to have lower addiction potential than heroin 2) Social settings were key facilitators of alcohol and methamphetamine use 3) Some participants, but not all, used methamphetamine to help quit heroin 4) Consuming alcohol blunted the effects of heroin, while paradoxically serving as a catalyst for heroin use 5) Use of methamphetamine was perceived by many participants to be incompatible with treatment for HIV. Conclusions Participant experiences reflected a significant impact of polysubstance use on treatment of HIV and OUD. Patterns of polysubstance use are subject to common preconceptions of alcohol and methamphetamine as having a low addictive potential, and these substances are deeply enmeshed in the social life of many people with OUD in Vietnam. Interventions to address complex social norms and potential harms of polysubstance use are urgently needed as the population of people receiving medication for OUD (MOUD) increases in Vietnam and globally. Trial registration BRAVO - NCT01936857, September 2013.

Background: Patients report that familial support can facilitate initiation and maintenance of antiretroviral therapy (ART) and medications for opioid use disorder (MOUD). However, providing such support can create pressure and additional burdens for families of people with opioid use disorder (OUD) and HIV. We examined perspectives of people with HIV receiving treatment for OUD in Vietnam and their family members. Methods: Between 2015 and 2018, we conducted face-to-face qualitative interviews with 44 patients and 30 of their family members in Hanoi, Vietnam. Participants were people living with HIV and OUD enrolled in the BRAVO study comparing HIV clinic-based buprenorphine with referral to methadone treatment at 4 HIV clinics and their immediate family members (spouses or parents). Interviews were professionally transcribed, coded in Vietnamese, and analyzed using a semantic, inductive approach to qualitative thematic analysis. Results: Family members of people with OUD and HIV in Vietnam reported financially and emotionally supporting MOUD initiation and maintenance as well as actively participating in treatment. Family members described the burdens of supporting patients during opioid use, including financial costs and secondary stigma. Conclusions: Describing the role of family support in the lives of people living with OUD and HIV in the context of Vietnam enriches our understanding of their experiences and will support future treatment efforts targeting the family unit.

RGS radiosurgery provided very high rates of symptom relief in patients with cerebral cavernomas.

Brainstem gliomas account for about 20% of primary brain tumors in children and for <2% of gliomas in adult patients (1-3). Gliomas of the brainstem are most commonly located in the pons, but are also found in the mesencephalon, the cerebellar peduncles or the medulla oblongata (4). A tumor located in the brainstem can be associated with severe clinical or even life-threatening symptoms, since this part of the brain regulates motion functions, as well as respiratory and circulatory functions. Such tumors include low-and high-grade lesions. The median survival time for patients with a brainstem glioma is about 10 months in children and 30-40 months in adult patients (5). The prognosis depends on the grade according to the World Health Organization (WHO) (6). In a retrospective study from the United States, median survival was 77 months for patients with WHO grade II lesions, 21 months for those with grade III lesions and 15 months for those with grade IV lesions (5). A great amount of research has been carried out on the treatment of higher grade (III or IV) lesions, whereas fewer data are available for patients with low-grade (grade II) gliomas (7-11). Therefore, the present study focused on low-grade gliomas of the brainstem and aimed to provide additional data for this less common situation.Since complete surgical resection is not safely possible in many of these patients, radiotherapy is the most common treatment. In order to allow for optimal sparing of normal tissues and structures surrounding the tumor, radiotherapy of brainstem gliomas should ideally be performed with highprecision techniques such as intensity-modulated radiotherapy, fractionated stereotactic radiation therapy or stereotactic radiosurgery (SRS). SRS is defined as singlefraction treatment, mostly performed with a linear accelerator or a gamma system. Two gamma systems are used for radiosurgery of cerebral lesions, including those located in the brainstem. The classic Gamma-Knife was already introduced about 50 years ago by Lars Leksell (12). The other system, the rotating gamma system (RGS), was developed in the late 1990s (13). The latter system includes technical features of the classic Gamma-Knife (Cobalt-60 sources) and linear acceleratorbased radiosurgery (a rotating gantry). Although radiosurgery with a gamma system appears to be a reasonable option for 957

Non - communicable diseases (NCDs) are the leading causes of death and disability globally. In Vietnam, NCDs account for 77% of all deaths. The purpose of this systematic review is to describe the patterns of mortality associated with NCDs in Vietnam from 2006 to 2016. We included 25 studies that reported on health impacts especially on mortality NCDs in Vietnam. From 2008 to 2012, NCDs deaths decreased by 2% (75% vs 73%) and from 2012 to 2016, it increased by 4% (73% vs 77%). The probability of premature deaths of NCDs in 2010, 2012 and 2016 were at 17.5%, 17.4% and 17.0%, respectively. Though the mortality patterns seem to decline over the years, it is not significant enough to conclude, and the probability of dying prematurely is still very high in the country. The percentage of deaths for male was 26.4% (2008), 54.3% (2012), 23.0% (2016) higher than those for female 19.4% (2008), 30.0% (2012), 11.0% (2016). Worldwide, 41 millions people die from non-communicable diseases each year with 15 millions people die between 30 - 69. Cardiovascular diseases remained the main leading cause of NCDs deaths, cancer was the second leading cause of NCDs deaths and projected to increase further in subsequent reports.

This paper aims to determine the rate of epidermal growth factor receptor (EGFR) gene mutations in advanced and metastatic squamous cell lung cancer in Vietnamese patients. EGFR mutation testing was performed on 559 formalin-fixed paraffin-embedded tissue samples from advanced and metastatic squamous cell lung cancer patients by real-time PCR. The results showed that EGFR mutations were found in 8.4%. Most of them were exon 19 deletions, approximately 51.1%. Multivariate analysis showed that EGFR mutations were more prevalent in gender (p<0.0001). In conclusion, EGFR mutation testing should be performed in patients with squamous cell lung cancer. Keywords: Lung squamous cell carcinoma, EGFR gene.