Professional Documents
Culture Documents
PDF Advances in Visual Informatics 6Th International Visual Informatics Conference Ivic 2019 Bangi Malaysia November 19 21 2019 Proceedings Halimah Badioze Zaman Ebook Full Chapter
PDF Advances in Visual Informatics 6Th International Visual Informatics Conference Ivic 2019 Bangi Malaysia November 19 21 2019 Proceedings Halimah Badioze Zaman Ebook Full Chapter
https://textbookfull.com/product/advances-in-visual-
computing-13th-international-symposium-isvc-2018-las-vegas-nv-
usa-november-19-21-2018-proceedings-george-bebis/
https://textbookfull.com/product/perspectives-in-business-
informatics-research-18th-international-conference-
bir-2019-katowice-poland-september-23-25-2019-proceedings-
Advances in Visual Computing 14th International
Symposium on Visual Computing ISVC 2019 Lake Tahoe NV
USA October 7 9 2019 Proceedings Part I George Bebis
https://textbookfull.com/product/advances-in-visual-
computing-14th-international-symposium-on-visual-computing-
isvc-2019-lake-tahoe-nv-usa-october-7-9-2019-proceedings-part-i-
george-bebis/
https://textbookfull.com/product/advances-in-visual-
computing-14th-international-symposium-on-visual-computing-
isvc-2019-lake-tahoe-nv-usa-october-7-9-2019-proceedings-part-ii-
george-bebis/
https://textbookfull.com/product/proceedings-of-the-
international-conference-on-advanced-intelligent-systems-and-
informatics-2019-aboul-ella-hassanien/
https://textbookfull.com/product/advanced-data-mining-and-
applications-15th-international-conference-adma-2019-dalian-
china-november-21-23-2019-proceedings-jianxin-li/
Advances in
Visual Informatics
6th International Visual Informatics Conference, IVIC 2019
Bangi, Malaysia, November 19–21, 2019
Proceedings
Lecture Notes in Computer Science 11870
Founding Editors
Gerhard Goos
Karlsruhe Institute of Technology, Karlsruhe, Germany
Juris Hartmanis
Cornell University, Ithaca, NY, USA
Advances in
Visual Informatics
6th International Visual Informatics Conference, IVIC 2019
Bangi, Malaysia, November 19–21, 2019
Proceedings
123
Editors
Halimah Badioze Zaman Alan F. Smeaton
Universiti Kebangsaan Malaysia Dublin City University
Bangi, Malaysia Dublin, Ireland
Timothy K. Shih Sergio Velastin
National Central University Carlos III University of Madrid
Taoyuan City, Taiwan Madrid, Spain
Tada Terutoshi Nazlena Mohamad Ali
Toyo University Universiti Kebangsaan Malaysia
Tokyo, Japan Bangi, Malaysia
Mohammad Nazir Ahmad
Universiti Kebangsaan Malaysia
Bangi, Malaysia
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
The world is going through great economic and technological challenges in this digital
economy era. With the advent of the 4th Industrial Revolution (4IR) and Industry 4.0,
Visual Informatics has become a significant component in computer science. This
multidisciplinary field encompasses that of computer science and information
technology and engineering, integrating areas such as computer vision, image
processing, pattern recognition, computer graphics, simulation, virtual reality, data
visualization and analytics, cyber security, as well as social computing, which can all
be applied to various domains of knowledge such as education, medical and health,
finance, or security. The Institute of Visual Informatics (IVI) at the Universiti
Kebangsaan Malaysia (UKM), also known as The National University of Malaysia, is a
center of excellence (CoE) established as an outcome of the first Visual Informatics
Conference (IVIC) held in 2009. The institute, which conducts research in the basic
areas mentioned earlier, conducts Masters and Doctoral (PhD) Degree programs of
research, as well as short Professional Practical Certifications currently in the areas of
Data Science and Intelligent Visual Data Analytics. The institute has successfully
graduated more than 30 postgraduate students (Masters and PhD) since its inception in
2010. We are indeed indebted to the international fraternity from the last five IVIC
conferences (2009, 2011, 2013, 2015, and 2017), who have given us support for the
establishment of the institute. Our smart partnerships, through this conference, has
helped the institute to grow and enhance research in this area locally and
internationally.
The Visual Informatics Research Group and the Institute of Visual Informatics
(IVI) at UKM, once again hosted this 6th International Visual Informatics Conference
(IVIC 2019), with the objective to bring together experts to discuss and share new
knowledge and ideas in this research area so that more concerted efforts can be
undertaken nationally and globally. As in the previous IVIC conferences, this
conference was organized collaboratively by the Visual Informatics fraternity from
various public and private universities and industry from various parts of the world
(their names are listed in the proceedings). This sixth edition of IVIC was co-sponsored
by Malaysian Digital Economy Corporation (MDEC), National Council of Professors
(MPN), and Malaysian Information Technology Society (MITS). The conference was
co-chaired by five professors from the UK, Ireland, Taiwan, Japan, and Malaysia.
The theme of the conference, ‘Digital Innovation Towards Society 5.0’ reflects the
importance of digitalization in this fourth industrial revolution digital economy, where
disruptive technologies have caused technologies to integrate and result in fusion
technologies that bring about interesting emerging technologies such as autonomous
and semi-autonomous vehicles, Internet of things (IoT), and Blockchain that will
benefit society. The changing landscapes of Industry 4.0 require technologies that are
Artificial Intelligence (AI)-driven which will assist future Society 5.0 to make decisions
from more precise intelligent predictions that would result in an efficient decision
vi Preface
making and smart intelligent applications for the well-being of society and the
economy. Thus, the theme of the conference was relevant, apt, and timely.
The conference focused on four tracks:
– Visualization and Digital Innovation for Society 5.0
– Engineering and Digital Innovation for Society 5.0
– Cyber Security and Digital Innovation for Society 5.0
– Social Informatics and Application for Society 5.0
The tracks took place over the course of two days (November 19–20, 2019) and
ended with a one-day workshop (November 21, 2019). There were 4 keynote speakers
and 65 paper presentations based on topics covered by the 4 main tracks. The
reviewing of the papers was conducted by experts who represented the Program
Committee from Asia, Europe, Oceania, and the USA. Each paper was reviewed by 3
reviewers and the acceptance rate was 47%. The reviewing process was managed using
the EasyChair system.
The conference also included for the first time an exhibition portraying the theme
‘Symbiosis of Art, Science, and Technology for Society 5.0’ to show how scientific
and technological research can be presented in interesting art works that have a great
impact on public awareness on the seamlessness of knowledge in art, science, and
technology, and its effects on the civilization of societies.
On behalf of the organizing and Program Committee of IVIC 2019, we thank all
authors for their submissions and camera-ready copies of papers, and all participants
for their thought-provoking ideas and active participation in the conference. We also
thank the vice-chancellor of UKM (host university), and vice-chancellors and deans of
all IT faculties of the IHLs for their support in organizing this conference. We also
acknowledge the sponsors, members of the Organizing Committees, Program
Committee members, support committees, and individuals who gave their continuous
help and support in making the conference a success. It is our fervent hope that one day
IVIC will be held in different host countries such as Asia, Europe, Oceania, the UK or
the USA.
The 6th International Visual Informatics Conference (IVIC 2019) was organized by the
Visual Informatics Research Group and Institute of Visual Informatics, Universiti
Kebangsaan Malaysia (UKM), in collaboration with local public and private
universities in Malaysia, the Multimedia Development Corporation (MDEC), and the
ICT Cluster of the National Professors’ Council (MPN).
Deputy Chair
Rosalina Abdul Salam (USIM)
Secretary
Nazlena Mohamad Ali (UKM)
Assistant Secretary
Norshita Mat Nayan (UKM)
Treasurer
Rabiah Abd. Kadir (UKM)
Assistant Treasurer
Ummul Hanan Mohamad (UKM)
Program Committee
Program Co-Chairs
Halimah Badioze Zaman Universiti Kebangsaan Malaysia, Malaysia
Alan F. Smeaton Dublin City University, Ireland
Timothy K. Shih National Central University, Taiwan
Sergio Velastin Queen Mary Univ. of London, UK
Terutoshi Tada Toyo University, Japan
viii Organization
Sponsorship
Azlina Ahmad (UKM) – Head
Halimah Badioze Zaman (UKM)
Rosalina Abdul Salam (USIM)
Wan Fatimah Wan Ahmad (UTP)
Bahari Belaton (USM)
Noor Afiza Mat Razali (UPNM)
Logistics
Riza Sulaiman (UKM) – Head
Mohamad Taha Ijab (UKM)
Syed Nasir Syed Zakaria Alsagoff (UPNM)
Dahlan Abdul Ghani (UNIKL)
Asama Kuder (UKM)
Azreen Azman (UPM)
Hafizhah Suzana Hussien (UKM)
Mohammad Haziq Mohd Johari (UKM)
Organization ix
Workshop
Puteri Nor Ellyza Nohuddin (UKM) – Head
Amelia Ritahani Ismail (UIA)
Nur Intan Raihana (USM)
Joshua Thomas (KDU)
Tour
Nurul Amelina Nasharuddin (UPM) – Head
Hajah Norasiken Bakar (UTEM)
Azreen Azman (UPM)
Exhibition
Mohamad Taha Ijab (UKM) – Head
Hanif Baharin (UKM)
Mohammad Haziq Mohd Johari (UKM)
Zulkepli Mukhtar (UKM)
Special Task
Norshita Mat Nayan (UKM) – Head
Siti NorazimahAhmat (UKM)
Nurulhuda Muhamad (UKM)
Anis Najwa Mohd Sharuddin (UKM)
Technical Committee
International
Alan F. Smeaton Dublin City University, Ireland
Timothy K. Shih National Central University, Taiwan
Sergio Velastin Queen Mary University of London, UK
Tada Terutoshi Toyo University, Japan
Emanuele Trucco University of Dundee, UK
Hang-Bong Kang Catholic University of Korea, South Korea
Marta Fairén Universitat Politècnica de Catalunya, Spain
Erich Neuhold University of Vienna, Austria
Theng Yin Leng Nanyang Technological University, Singapore
x Organization
Malaysia
Azlina Ahmad (UKM) Mahadi Bahari (UTM)
Aliimran Nordin (UKM) Maizatul H. M. Yatim (UPSI)
Amalina Farhi Ahmad Fadziah (UPNM) Marwan D. Saleh (MSU)
Asama Kuder Nseaf (UKM) Mohamad Taha Ijab (UKM)
Aslina Baharum (UMS) Mohammad Nazir Ahmad (UKM)
Azreen Azman (UPM) Mohammad Adib Khairuddin (UPNM)
Ang Mei Choo (UKM) Mohd Afizi Mohd Shukran (UPNM)
Bahari Belaton (USM) Mohd Hafiz Faizal Mohamad
Bavani Ramayah (NOTTINGHAM) Kamil (UNIKL)
Chiung Ching Ho (MMU) Mohd Nadhir Ab Wahab (USM)
Dahlan Abdul Ghani (UNIKL) Mohd Nizam Husen (UNIKL)
Dayang Rohaya Awang Rambli (UTP) Mohd Rizal Mohd Isa (UPNM)
Ely Salwana Mat Surin (UKM) Mohd Nazri Ismail (UPNM)
Falah Y. H. Ahmed (MSU) Muhammad Reza Z’aba (UM)
Fauziah Zainuddin (UMP) Muslihah Wook (UPNM)
Halimah Badioze Zaman (UKM) Nor Hidayati Zakaria (UTM)
Hanif Baharin (UKM) Nazlena Mohamad Ali (UKM)
Hajah Norasiken Bakar (UTEM) Nazmona Mat Ali (UTM)
Hoo Meei Hao (UTAR) Noor Afiza Mat Razali (UPNM)
J. Joshua Thomas (KDU Penang) Noor Azah Samsudin (UTHM)
Jamaiah Yahaya (UKM) Noor Hafizah Hassan (UTM)
Kasturi Dewi Varathan (UM) Noor Hayani Abd Rahim (IIUM)
Khairul Shafee Kalid (UTP) Noorminshah Iahad (UTM)
Organization xi
Strategic Partners
National Council of Professors (MPN)
Malaysia Digital Economy Corporation (MDEC)
Malaysian Information Technology Society (MITS)
Sistem Perintis Sdn Bhd
Co-organizers
Intrinsic Face Image Decomposition from RGB Images with Depth Cues. . . . 149
Shirui Liu, Hamid A. Jalab, and Zhen Dai
Virtual Reality Ideal Learning: A Look into Future Medical Classroom . . . . . 182
Mohd Yazid Bajuri, Kamarul Syariza Zamri, Mohammad Nazir Ahmad,
and Mohd Ismawira Mohd Ismail
Virtual Lower Limb Stroke Rehabilitation to Assess Post Stroke Patients . . . . 343
Lee Wei Jian and Syadiah Nor Wan Shamsuddin
1 Introduction
The World Health Organization (WHO) reports that around 253 million people have
some form of vision impairment and 36 million people are blind. These populations
would benefit from navigation assistance to help them in their daily lives. In recent
years, Computer Vision (CV) has created opportunities to develop systems that can
enable the visually impaired to connect with, recognize, and search their surroundings
[1]. However, building such navigation tools is difficult, especially in complex situa-
tions. Many methods have already been tried, including inexpensive canes, guide dogs,
and electronic travel aids. The difficulties that all of these solutions are trying to solve
can be classified into three categories: identifying the position of the person, the per-
son’s desired destination, and the best way to get to that destination [2]. Traditionally,
the blind and visually impaired have used a cane-like stick to detect object/obstacles in
the way. But now, more advanced technology has led to many research papers about
smart sticks that can classify object/obstacles via embedded sensors [3]. Researchers
have even added GPS into these walking sticks to obtain accurate location information,
and ultrasound to detect obstacles and potholes when the user is walking indoors or
outdoors [4]. Another navigation solution uses a virtual dog [5], which is implemented
by combining AlexNet and GoogleNet network for activity recognition. It can help the
blind and visually impaired to “see” using audio output from activity recognition.
We propose a new way to help the visually impaired navigate their surroundings
through image classification, object detection, and segmentation, which are imple-
mented via deep learning methods. We employ Mask-RCNN, a well-known model that
has been widely used to handle such issues. It has already been successfully applied to
oral disease detection and segmentation [6]. Moreover, its model can automatically
segment nucleic using a microscopic image. A navigation system must be able to
perform detection, localization and instance segmentation of a natural image [7]. Mask-
RCNN has been applied to hand segmentation using the COCO dataset and combined
with Mean Shift to improve the tracking results. Moreover, it can be applied to
dynamic hand segmentation for even better accuracy [8].
Using artificial intelligence, computer vision, and pattern recognition for classifi-
cation, tracking, detection, and recognition can be handled by Mask-RCNN. We want
to leverage the advantages of this method to build content-aware video analysis for
blind and visually impaired people to classify, track, detect, and recognize
object/obstacles around them. We use the COCO dataset, which contains many classes
of object/obstacles, for training.
We improve upon the Mask-RCNN results with the distance between the person
and object, additionally calculating the object’s speed and direction. We obtain the
class and score of an object from the Mask-RCNN results and find the distance by
using the threshold of the bounding box perimeter between the person and object using
two conditions (“medium” and “near”) in our system. We then calculate the total
distance and frame to get the speed of the object. The direction of the object is found
using the origin coordinates, point, and angle, which automatically show the direction.
Our contributions are thus threefold. (i) Calculate the distance between the subject and
the object; (ii) Calculate the object’s speed; (iii) Identify the direction of the object.
The paper is organized as follows. In Sect. 2, we present related literature. In
Sect. 3, we describe the Mask-RCNN method and our proposed method. In Sect. 4, we
present the experimental results. We draw conclusions and propose future work in
Sect. 5.
2 Related Work
The rapid global development of technology has spurred innovative assistive tech-
nologies, especially for the blind and visually impaired populations. For instance,
recently developed technology can help blind people easily identify clothes using NFC
(Near Field Communication) [9]. This system was developed on a Web platform, and
clothing characteristics are input manually. Blind people can also recognize products
for daily needs using a camera [10] since combining K-means clustering and SIFT
method allows for the separation of background and foreground, as well as recognition
of product labels.
Content-Aware Video Analysis 5
Today, deep learning has proven to be more powerful than machine learning
methods, especially in classification, tracking, detection, and recognition tasks [6, 7].
A deep neural network schema has been used to predict the future short-term speed
with a dataset consisting of a vehicle’s historic speed and acceleration, steering
information, area and driving time [11]. Positive results have shown promise for deep
learning methods. New deep learning models for object recognition (Faster-RCNN, R-
FCN, Mask-RCNN) and image feature extraction (InceptionResnet and Resnet) rely on
simple designs with better results. Mask-RCNN can be applied to synthetic images
only with the feature extractor frozen [12].
The latest event-recognition structure using deep local flow in fast region-based
convolutional neural network (Fast RCNN) is related to traffic violation recording,
monitoring, and control. The definition of the event varies in many aspects, but at its
core, an event in video detection is an object that manifests some behavior in the
scenes. So, in [13] event-recognition in the video consists of direction and speed of
object recognition, detection, and behavior. Content awareness also uses deep object
co-segmentation that segments a similar object from various images [14]. The work
compares three architectures, including channel-wise attention (CA), fused channel-
wise attention (FCA), and channel spatial attention (CSA) for classification between a
seen object and unseen object. Content awareness can additionally be implemented on
sports data to become a reference in context-aware research, especially sports videos
with respect to context-oriented groups, objects, and events [15].
3 Proposed Method
The general framework of our proposed method is shown in Fig. 1. It consists of four
main blocks. First, Mask-RCNN outputs name class and object score. Second, the
distance between the person and object is found using the bounding box perimeter
threshold. Third, the speed of the object is calculated using distance and frame total.
Lastly, we determine the direction of the object.
Fig. 3. Selected portions of a frame. Original image (left); after processing (right)
3.1 Mask-RCNN
Mask-RCNN is a popular deep learning model, especially for instance segmentation
[16]. It is derived from Faster RCNN and adds a mask for predicting segmentation on
every Region of Interest (RoI). The bounding box regression and existing branch
classification run in parallel. The technique applies a similar two-step procedure, with
an identical first stage, (which is RPN). The framework of Mask-RCNN is shown in
Fig. 2 [17]. For this research, we only take the important parts of an image to be
processed by Mask-RCNN. Those parts remain in RGB video, while other parts are
converted into the black by setting a value limit for the distance from the left and right
sides as shown in Fig. 3.
Here, we have two conditions for the distance: “near” and “medium”. “Near” means the
perimeter value exceeds 400. “Medium” means the perimeter value is between 300 and
400. This is denoted by Eq. (2).
Content-Aware Video Analysis 7
300\perimeter\400; medium
distance ð2Þ
perimeter [ 400; near
We have a perspective on the distance of an object. If an object is far from the person,
the size of an object is small (the perimeter is small). In turn, if an object is close to the
person, the object’s size is large (the perimeter is large) as shown in Fig. 4.
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
s¼ ð x1 x2 Þ 2 þ ð y1 y2 Þ 2 ð3Þ
distance ðsÞ
speed ðvÞ ¼ ð4Þ
time ðtÞ
3.4 Direction
The direction of a moving object is based on the center point, destination point, and
angle. The destination point is the same as the center point, and we add speed value
inside it so the length of the direction depends on the speed. We can find the angle by
calculating two coordinates (x, y) and then incorporating them with the cotangent
(tan−1) as shown in Eq. (5).
b 1
h ¼ tan ð5Þ
a
After we get the theta (h), we insert it into a trigonometric function that rotates a point
counterclockwise by an angle around the origin, which must be in radians. The
computation of the rotation is shown in Eq. (6), and the description of h is shown in
Fig. 6. R represents rotation, (i, j) represents the destination coordinates, (a, b) refers to
the center coordinates, and h is the angle.
x ¼ a þ cosðhÞ x ðiaÞ sinðhÞ x ðjbÞ
R ð6Þ
y ¼ b þ sinðhÞ x ðiaÞ þ cosðhÞ x ðjbÞ
Content-Aware Video Analysis 9
4 Experimental Results
4.1 Datasets
The dataset uses a common object in context (COCO). COCO is the largest image
dataset for object detection and segmentation purposes and can be applied in many
platforms, including MATLAB, Python, and Lua APIs. The COCO 2017 dataset
contains 123.287 images and 886.284 instances. We use this dataset without any edits.
software used Python 3, Keras, and TensorFlow. Furthermore, we had to install another
library that included numpy, scipy, pillow, cython, matplotlib, scikit-image, opencv-
pyhton, h5py, imgaug, and Ipython. Afterward, we could run the Mask-RCNN pro-
gram to obtain outputs for the name of the label class, score, mask and bounding box.
The original results of Mask-RCNN are shown in Fig. 7.
The mind is generally clear in the early stage; afterward there may
be delirium, especially along with cerebral complication.
MORBID ANATOMY.—It is rarely that the pia mater is the only tissue
involved in the inflammation. Frequently the inner surface of the dura
is the seat of a fine injection, with delicate false membranes, and the
cord itself—at least its exterior portions—probably always
participates more or less in the congestion. The pia is reddened and
thickened, the surface showing small bloody extravasations, and the
space between its two layers is the seat of a fibro-purulent deposit.
The spinal fluid is turbid and flocculent. The seat and extent of the
morbid appearances vary in different cases; they are always more
abundant in the posterior than the anterior part of the cord, and may
be confined to a limited space or extend throughout its whole length.
It is remarkable that the region of the medulla oblongata is generally
free or only slightly affected; but since bulbar symptoms are often
prominent in grave cases, Leyden6 accounts for it by supposing that
the exudation is washed away by the constant movement of the
cerebro-spinal fluid. If the cord be involved in the inflammation, it is
softened and injected, the nerve-sheaths are destroyed, and the
axis-cylinders swollen in places. The nerve-roots show hyperæmia,
infiltration of the interstitial tissue with round cells, and destruction of
the nerve-sheaths.
6 Klinik der Rückenmarks-krankheiten, von E. Leyden, Berlin, 1874, vol. i. p. 407.
Chronic spinal meningitis may follow the acute form, or it may arise
from chronic disease of the vertebræ or of the cord, especially
myelitis and sclerosis. It is most apt to accompany sclerosis of the
posterior columns, and it is often difficult to say in any particular case
whether the meningeal affection preceded or followed that of the
cord. Probably some cases of chronic myelitis, especially of the
disseminated form, owe their origin to chronic meningitis.7 It has
been thought to follow blows on the back, and also to arise from
general concussion without traumatism, and has been considered as
a frequent result of accidents from railroad collisions, etc. This view
has been disputed by Herbert W. Page,8 who says: “Of the
exceeding rarity of spinal meningitis as an immediate result of
localized injury to the vertebral column we are well assured.... And
we know of no one case, either in our own experience or in the
experience of others, in which meningeal inflammation has been
indisputably caused by injury to some part of the body remote from
the vertebral column.” Chronic alcoholism and syphilis, especially the
latter, predispose to the disease. In many cases no adequate cause
can be assigned.
7 Leyden, op. cit., vol. i. p. 442.
8 Injuries of the Spine and Spinal Cord, without Apparent Mechanical Lesion and
Nervous Shock, in their Medico-legal Aspects, London, 1883, p. 128.
13 A case of scorbutic spinal hemorrhage is reported in the British Med. Journal, Nov.
19, 1881.
As may be inferred from what has already been said, the course of
spina bifida is usually rapid and toward a fatal termination. In some
instances, however, as in cases observed by Holmes, and more
recently by Lithgow, spontaneous recovery has followed the
obliteration of the channel which unites the sac with the cavity of the
spinal membranes; and in other instances, without a cure having
been effected, life has been prolonged for very many years. Thus,
Behrend reports a case in which a patient with spina bifida lived to
the age of fifty, and Holmes refers to another in which death resulted
from an independent disease at the age of forty-three. But a still
more remarkable case was recorded by Callender, the patient in this
instance having reached the age of seventy-four.
The introduction of iodine into the sac of a spina bifida is, according
to Morton, only justifiable in cases unattended by paralysis; under
opposite circumstances I should be disposed to try a plan recently
employed with success by Noble Smith in a case of meningocele—
viz. injecting the iodo-glycerin solution into the coverings of the sac,
and as close to it as possible without perforating it.
(3) and (4). Ligation and excision have each occasionally effected a
cure, but more often have but helped to precipitate a fatal issue. A
successful case of ligation followed by excision has been recently
recorded by Löbker. The elastic ligature, applied around the neck of
the sac (if this be pedunculated), has been employed by Laroyenne,
Ball, Colognese, Baldossare, Mouchet, and other surgeons, and of 6