=Paper=
{{Paper
|id=Vol-2544/shortpaper3
|storemode=property
|title=A Mobile Online Platform for Aged Men’s Prostate Hypertrophy Monitoring Based on Linea Nigra Images Analysis
|pdfUrl=https://ceur-ws.org/Vol-2544/shortpaper3.pdf
|volume=Vol-2544
|authors=Géraud M. Azehoun-Pazou,Gloria Tognon Tchegnonsi,Bickel Oloude,Kokou M. Assogba,Hugues Adegbidi
|dblpUrl=https://dblp.org/rec/conf/irehi/Azehoun-PazouTO18
}}
==A Mobile Online Platform for Aged Men’s Prostate Hypertrophy Monitoring Based on Linea Nigra Images Analysis==
https://ceur-ws.org/Vol-2544/shortpaper3.pdf
A Mobile Online Platform for Aged Men’s Prostate
Hypertrophy Monitoring Based on Linea Nigra
Images Analysis
Mahugnon Géraud
Kokou Assogba Hugues Adegbidi Gloria Tognon Bickel Oloude
Azehoun-Pazou
Laboratory of Faculty of Health Tchegnonsi Tamakan Inc.,
National University of Electrotechnic, Sciences, University 07 BP 265, Benin
Tamakan Inc.,
Science, Technology, telecommunication of Abomey-Calavi, bickeloloude@gmail.c
07 BP 265, Benin
Engineering and and Applied 01 BP 526, Benin om
tchegnonsigloria@gm
Mathematics, Benin Informatics, Benin adegbidih@yahoo.fr ail.com
geraud.pazou@unstim mkokouassogba@yah
.bj oo.fr
Abstract—Aged men generally have a hypertrophied prostate. affect health. In particular, image processing algorithms can
This hypertrophy can evolve in cancer or can just be a benign provide quantification, accuracy, reliability and repeatability of
hyperplasia. Some recent studies revealed that a linear measurements and analyzes by delegating tasks to computer.
hyperpigmentation of skin called Linea nigra (LN), can appear in Because of that, development of computer aided diagnostic
men developing prostate cancer or benign prostatic hyperplasia. systems (CADS) is a research field of interest [4]. Their goal is
LN’s prevalence is higher in the case of prostate cancer. It has to help physicians to make their decision through image
been established that when a LN appears on someone’s analysis. Despite these advances, there are other challenges that
suprapubic region and has a length greater than ten centimeters, dermatologist have to face. Indeed, in poor countries, there are
it indicates a metastasized or metastasizing tumor. Based on this,
very few specialists to cover a large population. This is the case
we developed in a previous work a method of LN segmentation.
From segmented images, we can calculate descriptive values such
in Mali where there is less than one dermatologist per million
as length, width, area, texture and color. These values are LN inhabitants. As a result, the country has to deal with a low level
characteristics that can be used by specialists to make diagnosis. of competence in peripheral health structures. In such context,
In this paper, we propose an implementation of those methods in portable applications which are close to patients (including
an Android mobile application. The aim is to connect patients to telemedicine, mainly tele-expertise) could be the solutions.
their specialists for distant monitoring purpose based on images Thanks to recent advances in ICT, it is possible for devices
taken with their mobile phones. The specific architecture of the (mainly smart phones and tabs) integrating software and
platform also helps in providing more precision in detection applications to be able to make decisions about a dermal
thanks to a deep learning program. disease based on images of pigmented disorders.
Keywords— linea nigra images analysis; mobile application; Recent advances in computing and Artificial Intelligence
aged persons distant health monitoring (AI) have had a huge impact on interpretation of medical
images. However, since obtained results cannot be fully taken
I. INTRODUCTION into account, it is essential to have final conclusions from
Usually found in pregnant women, the linea nigra (LN) is a doctors. It is for all these reasons and also in order to ensure a
linear hyperpigmentation of skin that appears between navel fast, cost-effective and remote consultation for everyone that
and suprapubic region. It sometimes appears in men of a we propose in this paper an online platform whose aim is to
certain age and non-pregnant women. Generally, LN is found detect and to analyze pigmented disorders, while taking into
on the abdomen of about 75% of women in pregnancy [1]. account the expertise of distant specialists.
However, studies published in [2], [3] revealed that linea nigra II. EASE OF USE
is observed in men having prostate cancer or benign prostatic
hyperplasia, Ly et al. (2013) proved that the prevalence is Using mathematics and technological advances, scientists
higher in the case of prostate cancer and concluded in [2] that created different applications to equip dermatologists. These
an LN which is greater than ten centimeters indicates a applications also help to raise awareness and bring medical
metastatic tumor. From this conclusion, it appears that prowess closer to population. Among them, we can enumerate:
characterization of LN can be used as a non-invasive method to • A special skin cancer screening application developed by
establish a differential diagnosis between the two types of SkinVision that can be downloaded to smartphone [5]. It
prostatic tumors. This characterization can be done from detects potentially suspicious moles and skin lesions and then
images, thanks to image processing. records snapshots to detect abnormal changes. It allows
Medical images are one of the most widely used tools to photographing the suspect mole and then analyzes it in seconds
better understand both normal and abnormal processes that using an algorithm that identifies any potentially abnormal
growth. The application is the first comprehensive application
Copyright © 2019 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0)
IREHI 2018 : 2nd IEEE International Rural and Elderly Health Informatics Conference
of skin cancer that has achieved European Certification (EC). It research institutions in Europe specialized in skin cancer.
has been scientifically tested in 2013 by Munich University SkinVision’s application is used by in many european countries
(LMU) Clinic, which is recognized as one of leading academic
Fig 1. System architecture
• Maryam Sadeghi developed Molescope, an application
that, using a smart phone connected to a mini-microscope, B. Final users
allows to take very precise images of spots and moles in order People involved in using the platform are:
to detect possible malignant melanomas. [6]
Sick people;
• A team from University of Houston, USA, developed the
DermoScreen application, an application capable of preventing Population;
skin cancer. By photographing a suspect lesion or mole with Dermatologists (doctor).
the phone, the software analyzes the risks whether it is cancer,
with a success rate of 85%. [7] Generalist doctors
III. MATERIAL AND METHODS
C. System architecture
A. Methodology The system of LN detection application integrates in its
Applications conception mostly for mobiles have become for functioning the mobile application associated with web
some years accessible to the community; that is possible to the interface for computers used by everybody as he is a doctor or
presence and availability of numerous modeling and patient (figure 1). The network cloud is made up by images
programming tools. However, in the frame of medical processing and treatment algorithms implemented for LN
applications pointed, it is generally professionals we meet. evaluation.
The conceptual phase stands for to take into account every
features and mathematical treatments to do on the images for D. Networks architecture
LN evolution detection. The database modeling is edited with A mobile application running on smartphone and using the
UML. The development phase of platform must follow MCV network connectivity for rural populations mostly need to be
for ease the application maintenance. soft in its use. It means that processes and technologies used to
develop and deploy application must not be smartphone chip
gluttonous.
Access to internet is generally difficult in rural and remote Tchat OTT (mini OTT) module has two modes: mode
places. At the first side, the coverage of terrestrial mobile reserved (discussion group for one treated patient
networks is not enough because the mobile networks operators doctors) and mode open (discussion group between
judge useless to densify with base stations or allocate a great doctors college and a patient).
bandwidth for a land where mobile traffic is not important. At Patients can send LN images to the online platform
the other side, the bandwidth and data rate put available are from a mobile phone;
inappropriated for increasing rural needs. In front of this fact,
mobile API mustn’t be greedy in chip capacities and data. As soon as an image is sent by a patient, his doctor is
notified. From then on, doctor can connect to the
platform and launch image analysis process;
At the end of an analysis process, found characteristics
can be used by specialists to make diagnosis.
Sick patient receives advices and possibly prescriptions
or an appointment for a thorough auscultation.
G. Technical choices
For technical choices, we bet on python technologies.
Firstly, Python is multiplatform. Secondly, Python is one of
best scripting languages with supported libraries in Image
processing domain. The image processing module must be at
the basis entirely in Python (opencv, numpy, matplotlib...).
Thus, the details relative to mobile application using Python
technologies are:
Fig 2: Rural scenario of application
- Front-end: Kivy is the Python mobile API selected. Using
Kivy, mobile application done can be distributed on PlayStore.
E. General (standard) features The application can use also Android services (SMS, camera,
The mobile platform useful for remote people must integrate mail and notifications system); and have access to most of the
normal java API.
the basic functions like:
- Back-end: SL4A (Scripting Layer for Android). SL4A
makes possible scripting languages to Android by using Python
for complex processes.
- Web: The web interface accessible is guaranteed by
Django utilization web version of application in order to allow
to medical personnel to have flexibility in their technical works
H. Image processing module
The image processing module has been developed in
python programming language based on LN’s segmentation
algorithm proposed in [9]. It consists of the following steps:
conversion from color to grayscale based on Principal
Component Analysis results; second step is contour
initialization; in third step, contour evolves iteratively until it
no longer changes.
Once LN images has been segmented, some characteristic
Fig 3: Basic mobile features values can be calculated. They are: length, width, ratio, area,
and histogram of oriented gradients.
F. LN Application available features 1) Length and Width
The following features are available: They are determined from two furthest points considering
respectively X axis and Y axis.
Sick people (patients) can sign up (register) on the
platform. They can register with a generalist ID
associated or a dermatologist associated. The first
doctor associated to one patient must validate the
enrolment before the continuation (the patient will 2) Ratio
have access to health features). It is the ratio of width to length, gives an indication of the
Medical personnel must sign up with mandatory form of the pigmented disorder. Indeed, a ratio very close to
informations useful to track their service states; zero indicates a very sharp LN. A value of ratio close to 0.5
indicates a LN more or less wide and / or not very long. A ratio
The doctor treating a patient can add in need case till close to 1 indicates a regular shape: circle or square.
two specialists;
3) Area
It corresponds to the total number of pixels that belong to are calculated: Length = 365 pixels; Width = 50 pixels; Ratio =
segmented region. 0.137 and Area = 5191 pixels.
4) Histogram of oriented gradients (HOG) V. CONCLUSION
This descriptor calculates local histograms of the gradient
The development of a technology combining a large
orientation on a dense grid (evenly distributed areas) of the
amount of information on skin disorders and the analysis of
images. First proposed in [] the method has proven
data calculated from images can increase the information
effectiveness for the detection of people.
available for Dermatologists. This motivated our research on
the development of a platform to serve as a diagnostic aid
system. The developed platform is suitable for elderly people
IV. RESULTS whose mobility possibilities become reduced with advanced
age. It is also very useful for people living in rural areas.
In future work, we should focus on automatic detection of
probable or non-probable metastasis of a prostate tumor based
on identified characteristics combined with results of
biomedical analyzes and clinical symptoms.
REFERENCES
[1] E. Esteve, L., Saudeau, F., Pierre, K., Barruet, L., Vaillant, G., Lorette.
‘‘Physiological cutaneous signs in normal pregnancy: a study of 60
pregnant women, in: Annales de Dermatologie et de Venereologie’’., pp.
227–231, 1993.
[2] L. I. Okeke, A. O. George, A. O. Ogunbiyi, et M. Wachtel, ‘‘ Prévalence
de la Linea Nigra chez des patients souffrant d’Hyperplasie Bénigne de
la Prostate et de Cancer de la Prostate: Hyperplasie Bénigne de la
Prostate, Linea Nigra, Cancer de prostate ’’, Int. J. Dermatol., vol. 51, p.
45‑48, nov. 2012.
[3] F. Ly et al., ‘‘ Linea nigra et tumeurs de la prostate ’’, Ann. Dermatol.
Vénéréologie, vol. 140, no 12, p. S560, déc. 2013.
Fig4: Some LN segmentation results that Doctors can view
[4] K. Korotkov et R. Garcia, ‘‘ Computerized analysis of pigmented skin
lesions: a review ’’, Artif. Intell. Med., vol. 56, no 2, p. 69–90, 2012.
[5] C. Lallemand, C.. "Cancer de la peau: une application pour analyser les
grains de beauté suspects". Mise à jour le 06 Mai 2015. In LeVif
l’Express., [Online]. http://www.levif.be/actualite/sante/cancer-de-la-
peau-une-application-pour-analyser-les-grains-de-beaute-
suspects/article-normal-393261.html [consulté le 13 septembre 2017].
[6] F., Arnould. "Une application pour détecter les cancers de la peau". Mise
à jour le 14 Juin 2015. In Radio Canada., [Online]. http://ici.radio-
canada.ca/nouvelle/724701/telephone-intelligent-microscope-cancer-
peau [consulté le 11 septembre 2017].
[7] E., Bizzotto. "Cancer de la peau : une application pour analyser les
risques". Mise à jour le 13 Mai 2014. In Top Santé., [Online].
https://www.topsante.com/medecine/cancers/cancer-de-la-peau/cancer-
de-la-peau-une-application-pour-analyser-les-risques-58423 [consulté le
15 septembre 2017].
[8] G. Azehoun-Pazou, K. M. Assogba, & H. Adegbidi, “A Multispectral
Analysis of Black Skin Color Images for Linea Nigra Segmentation”. In
Fig5: Graph showing a segmented LN dimensions Proceedings of 2017 IEEE International Conference on Bio-engineering
for Smart Technologies (Biosmart 2017), pp. 27-30, 2017.
[9] Dalal N., & Triggs B. “Histograms of oriented gradients for human
Two segmentation results are shown in figure 4. Since it is detection”, in proceedings of 2005 IEE Conference on Computer Vision
important from clinical point of view to circumscribe the whole and Pattern Recognition (CVPR 2005). pp. 886–893, 2005.
NL, we can conclude that our method is good. Considering the
segmented LN of figure 5, the following characteristic values