Patient Medical Images Research Articles

Paradigm Shift in Rhinoplasty with Virtual 3D Surgery Software and 3D Printing Technology.

Most Asians have a nose with a short columella and a low dorsum; augmentation rhinoplasty using implants is commonly performed in Asian countries to achieve a taller and more well-defined nasal dorsum. However, the current knowledge is insufficient to fully understand the various subjective desires of patients, reflect on them during surgery, or to objectively analyze the results after surgery. Advances in digital imaging technologies, such as 3D printing and 3D scanning, have transformed the medical system from hospital-centric to patient-centric throughout the medical field. In this study, we applied these techniques to rhinoplasty. First, we used virtual 3D plastic surgery software to enable surgical planning through objectified numerical calculations based on the visualized data of the patient's medical images rather than simple virtual plastic surgery. Second, the customized nasal implant was manufactured by reflecting the patient's anatomical shape and virtual 3D plastic surgery data. Taken together, we describe the surgical results of applying these rhinoplasty solutions in four patients. Our experience indicates that high fidelity and patient satisfaction can be achieved by applying these techniques.

Computer-aided diagnosis systems: a comparative study of classical machine learning versus deep learning-based approaches.

The diagnostic phase of the treatment process is essential for patient guidance and follow-up. The accuracy and effectiveness of this phase can determine the life or death of a patient. For the same symptoms, different doctors may come up with different diagnoses whose treatments may, instead of curing a patient, be fatal. Machine learning (ML) brings new solutions to healthcare professionals to save time and optimize the appropriate diagnosis. ML is a data analysis method that automates the creation of analytical models and promotes predictive data. There are several ML models and algorithms that rely on features extracted from, for example, a patient's medical images to indicate whether a tumor is benign or malignant. The models differ in the way they operate and the method used to extract the discriminative features of the tumor. In this article, we review different ML models for tumor classification and COVID-19 infection to evaluate the different works. The computer-aided diagnosis (CAD) systems, which we referred to as classical, are based on accurate feature identification, usually performed manually or with other ML techniques that are not involved in classification. The deep learning-based CAD systems automatically perform the identification and extraction of discriminative features. The results show that the two types of DAC have quite close performances but the use of one or the other type depends on the datasets. Indeed, manual feature extraction is necessary when the size of the dataset is small; otherwise, deep learning is used.

Auto-Diagnosis of Lung Cancer with the Proposed Feature Fusion-Based Hybrid Deep Model

Early detection of lung cancer increases the response rate to treatment. Therefore, the accuracy of diagnostic methods is of great importance. Reading the patient's medical images by radiologists can cause a severe time cost besides subjective result. In this context, Artificial Intelligence (AI) methods create an innovative field to reduce the workforce of radiologists and obtain objective results. AI methods play a vital role in improving the analysis of the dataset, extracting meaningful features, clustering, and classification. In our study, the data set contains healthy images besides CT images of malignant and benign tumors with lung cancer; AlexNet is trained using DenseNet 201, GoogleNet, MobileNetV2, and ResNet50 architectures. In addition, a hybrid model has been developed to classify lung CT images. The developed model constitutively used GoogleNet, MobileNetV2, and ResNet50 architectures. The feature maps obtained in these three architectures were combined and classified into different classifiers. Among the classifiers used in the study, the highest accuracy rate was achieved in the Ensemble Subspace KNN classifier. The accuracy value obtained in this classifier is 98.3%.

B‐SCORE – A blockchain based hybrid chaotic signatures for medical image encryption in an IoT environment

AbstractInternet of things (IoT) has evolved exponentially in the recent years and its applications are also being explored in the medical field. Due to this, the volume of medical images transmitted has increased multifold. Usage of IoT networks for medical image transmission has significantly reduced the time needed for clinical diagnosis and thereby increasing treatment efficiency. However at present, IoT networks are open to various security threats, which may affect the sensitive and private information that are present in patient's medical image datasets. Existing studies reveal the need of improvisation for secured medical data transmission over IoT networks. In the context to IoT security issues, this research paper proposes blockchain architecture integrated with chaotic encrypted medical image transmission to ensure the high security in medical image transmission. The proposed system incorporates tri‐layered architecture such as Image Aware Segmentation (IAS), hybrid chaotic encryption scheme and finally blockchain environment. The extensive experimentation has been carried out in which the performance parameters such as entropy, NACI and UACI (Number of Pixel Change Ratio and Unified Average Changed Intensity) were calculated and analyzed. It is found that the proposed architecture has NPCR as 99.65%, UACI as 33.95% and entropy ideally close to 8. Encryption results show that the proposed architecture exhibited more randomness, which can defend the IoT security threats.

A secure cloud system for maintaining COVID-19 patient's data using image steganography

Background: The COVID-19 pandemic has been life-threatening for many people and as such, a contactless medical system is necessary to prevent the spread of the virus. Smart healthcare systems collect data from patients at one end and process the acquired data at the other end. The cloud is the central point and the communication happens through insecure channels. The main concern, in this case, is the violation of privacy and security as the channel is untrusted. Traditional methods do not provide enough hiding capacity, security, and robustness. This work proposes an image steganography method using the deep learning method to hide the patient's medical images inside an innocent cover image in such a way that they are not visible to human eyes which reduces the suspicions of the presence of sensitive data. Methods: An auto encoder-decoder-based model is proposed with three components: the pre-processing module, the embedding network, and the extraction network. Features from the cover image and the secret images are extracted and fused to reconstruct the stego image. The stego image is then used to extract the ingrained secret image.shows the overall system workflow. Results: Peak Signal-to-Noise Ratio (PSNR) is the evaluation metrics used. The ImageNet dataset was used for training and testing the proposed model.shows the image results of the proposed method. Conclusion: During a COVID-19 screening test, private patient data such as mobile number and Qatari identity card are collected, transferred, and stored through untrusted channels. It is of paramount importance to preserve the privacy, security, and confidentiality of the collected patient records. A secure deep learning-based image steganography method is proposed to secure the sensitive data transferred through untrusted channels in a cloud-based system.

Mobile-based Telemedicine Application using SVD and F-XoR Watermarking for Medical Images

A medical- service platform is a mobile application through which patients are provided with doctor’s diagnoses based on information gleaned from medical images. The content of these diagnostic results must not be illegitimately altered during transmission and must be returned to the correct patient. In this paper, we present a solution to these problems using blind, reversible, and fragile watermarking based on authentication of the host image. In our proposed algorithm, the binary version of the Bose_Chaudhuri_Hocquengham (BCH) code for patient medical report (PMR) and binary patient medical image (PMI) after fuzzy exclusive or (F-XoR) are used to produce the patient's unique mark using secret sharing schema (SSS). The patient’s unique mark is used later as a watermark to be embedded into host PMI using blind watermarking-based singular value decomposition (SVD) algorithm. This is a new solution that we also proposed to applying SVD into a blind watermarking image. Our algorithm preserves PMI content authentication during the transmission and PMR ownership to the patient for subsequently transmitting associated diagnosis to the correct patient via a mobile telemedicine application. The performance of experimental results is high compare to previous results, uses recovered watermarks demonstrating promising results in the tamper detection metrics and self-recovery capability, with 30db PSNR, NC value is 0.99.

A Dynamic Three-Bit Image Steganography Algorithm for Medical and e-Healthcare Systems

Massive advances in internet infrastructure are impacting e-healthcare services compared to conventional means. Therefore, extra care and protection is needed for extremely confidential patient medical records. With this intention, we have proposed an enhanced image steganography method, to improve imperceptibility and data hiding capacity of stego images. The proposed Image Region Decomposition (IRD) method, embeds more secret information with better imperceptibility, in patient's medical images. The algorithm decomposes the grayscale magnetic resonance imaging (MRI) images into three unique regions: low-intensity, medium-intensity, and high-intensity. Each region is made up of k number of pixels, and in each pixel we operate the block of n least significant bits (LSBs), where 1 ≤ n ≤ 3. Four classes of MRI images of different dimensions are used for embedding. Data with different volumes are used to test the images for imperceptibility and verified with quality factors. The proposed IRD algorithm is tested for performance, on the set of brain MRI images using peak signal-to-noise ratio (PSNR), mean square error (MSE) and structural similarity (SSIM) index. The results elucidated that the MRI stego image is imperceptible, like the original cover image by adjusting 2 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">nd</sup> and 1 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">st</sup> LSBs in the low-intensity region. Our proposed steganography technique provides a better average PSNR (49.27), than other similar methods. The empirical results show that the proposed IRD algorithm, significantly improves the imperceptibility and data embedding capacity, compared to the existing state-of-the-art methods.

Medical Images are Safe - an Enhanced Chaotic Scrambling Approach.

The patient data confidentiality is one of the vital security aspects in e-Health and m-Health services. In particular, providing confidentiality to the patient's medical image is essential and the protection approaches have to be explored in-depth due to the rapid progress in the technologies of teleradiology and PACS. In this study, the pseudo random number generators (PRNGs), namely, the linear congruential generator (LCG) and XOR shift generator (XSG) are improved and combined with improved logistic 2D coupled chaotic map to provide enhanced chaos based encryption. The proposed scheme encrypts the Digital Imaging and Communication in Medicine (DICOM) images to protect the patient confidentiality during the storage and transfer in radiological information system (RIS). The cipher image was measured with various security analyses and tested with different test suites to prove its randomness.

An optimisation method for fusion-based reversible embedding and retrieving in MRI/CT medical images

Embedding and retrieving have an extensive scope of applications in the medical field. This method is worthwhile in strengthening the particulars of patients from the trespasser with prime storage capacity. The medical images of different modalities along with Patient Medical Image (PMI) can be conveyed to the medical practitioner across the globe for the opinion. Owing to the bandwidth and storage restraint, medical images must be compressed prior to the transference and storage. This paper presents a novel method for fusion-based embedding and retrieving by using Modified Complex Contourlet Transform (MCCT) and Modified Grey Wolf Optimisation (MGWO). It also determines the compression method by using Complex Contourlet Transform (CCT). Hence, this paper suggests that the fusion-based embedding and retrieving technique based on MCCT and MGWO is providing better hiding capacity. The improved values of compression ratio (CR), space saving (SS), mean square error (MSE) and peak signal to noise ratio (PSNR) show that the new novel method fulfils the properties of the embedding and retrieving technique.

Cloud Image Data Center for Healthcare Network in Taiwan.

This paper investigates how a healthcare network in Taiwan uses a practical cloud image data center (CIDC) to communicate with its constituent hospital branches. A case study approach was used. The study was carried out in the central region of Taiwan, with four hospitals belonging to the Veterans Hospital healthcare network. The CIDC provides synchronous and asynchronous consultation among these branches. It provides storage, platforms, and services on demand to the hospitals. Any branch-client can pull up the patient's medical images from any hospital off this cloud. Patients can be examined at the branches, and the images and reports can be further evaluated by physicians in the main Taichung Veterans General Hospital (TVGH) to enhance the usage and efficiency of equipment in the various branches, thereby shortening the waiting time of patients. The performance of the CIDC over 5 years shows: (1) the total number of cross-hospital images accessed with CDC in the branches was 132,712; and (2) TVGH assisted the branches in keying in image reports using the CIDC 4,424 times; and (3) Implementation of the system has improved management, efficiency, speed and quality of care. Therefore, the results lead to the recommendation of continuing and expanding the cloud computing architecture to improve information sharing among branches in the healthcare network.

Region image sharing with various secrecy levels

In most secret image sharing schemes, the secrecy level of each pixel is the same. However, in real applications, different pixels may require different secrecy levels. For example, in a patient's medical image, the region with the patient's name must be highly protected. To solve this problem, Wang and Lin proposed a scheme that allows a dealer to divide an image into multiple regions, each of which is associated with a certain level of secrecy. In their scheme, an extra pixel-security-level map (PSLM) is required to record the secrecy level of each pixel, and the shadow size of the PSLM is approximately 20 % of that of the secret image. It is unreasonable to have such a large overhead. To reduce the shadow size of the PSLM, they restricted each region to be a rectangle. Furthermore, any two participants can reconstruct the PSLM, enabling the content of regions with high secrecy levels to be compromised easily. To resolve these disadvantages, this paper proposes a new encoding scheme for PSLMs. The location of each region is protected according to its secrecy level, thus improving the security of the image. Moreover, the shape of each region is unrestricted, and the shadow size of the PSLM is reduced significantly.

Knee kinetics and kinematics: What are the effects of TKA malconfigurations?

Total knee arthroplasty (TKA) is a very successful surgical procedure. However, implant failures and patient dissatisfaction still persist. Sometimes surgeons are not able to understand and explain these negative performances because the patient's medical images "look good", but the patient "feels bad". Apart from radiograph imaging and clinical outcome scores, conventionally used follow-up methods are mainly based on the analysis of knee kinematics. However, even if kinematics remains close to the "normal" range of motion, the patient may still complain about pain and functional limitations. To provide more insight into this paradox, a better quantitative understanding of TKA mechanics must be developed. For this purpose, improved techniques for clinical follow-up, combining kinetics and kinematics analysis, should be introduced to help surgeons to assess and understand TKA performance. An analysis on four TKA designs was performed, and the changes in kinematics and in kinetics induced by several implant configurations (simulating implant malalignment and different knee anatomy) were compared. More specifically, analysed tibio-femoral and patello-femoral contact forces and tibio-femoral kinematics were analysed during a squat task up to 120°. The results from this study show that contact forces (with changes up to 67%) are more heavily affected by malconfigurations than kinematics, for which maximum deviations are of the order of 5mm or 5°, similar to the simulated surgical errors. The results present a similar trend for the different designs. The results confirm the hypothesis that kinematics is not the only and also not the most relevant parameter to predict or explain knee function after TKA. In the future, techniques to analyse knee kinetics should be integrated in the clinical follow-up.

Real-time 3D image reconstruction guidance in liver resection surgery.

Minimally invasive surgery represents one of the main evolutions of surgical techniques. However, minimally invasive surgery adds difficulty that can be reduced through computer technology. From a patient's medical image [US, computed tomography (CT) or MRI], we have developed an Augmented Reality (AR) system that increases the surgeon's intraoperative vision by providing a virtual transparency of the patient. AR is based on two major processes: 3D modeling and visualization of anatomical or pathological structures appearing in the medical image, and the registration of this visualization onto the real patient. We have thus developed a new online service, named Visible Patient, providing efficient 3D modeling of patients. We have then developed several 3D visualization and surgical planning software tools to combine direct volume rendering and surface rendering. Finally, we have developed two registration techniques, one interactive and one automatic providing intraoperative augmented reality view. From January 2009 to June 2013, 769 clinical cases have been modeled by the Visible Patient service. Moreover, three clinical validations have been realized demonstrating the accuracy of 3D models and their great benefit, potentially increasing surgical eligibility in liver surgery (20% of cases). From these 3D models, more than 50 interactive AR-assisted surgical procedures have been realized illustrating the potential clinical benefit of such assistance to gain safety, but also current limits that automatic augmented reality will overcome. Virtual patient modeling should be mandatory for certain interventions that have now to be defined, such as liver surgery. Augmented reality is clearly the next step of the new surgical instrumentation but remains currently limited due to the complexity of organ deformations during surgery. Intraoperative medical imaging used in new generation of automated augmented reality should solve this issue thanks to the development of Hybrid OR.

Electroporation-Based Treatment Planning for Deep-Seated Tumors Based on Automatic Liver Segmentation of MRI Images

Electroporation is the phenomenon that occurs when a cell is exposed to a high electric field, which causes transient cell membrane permeabilization. A paramount electroporation-based application is electrochemotherapy, which is performed by delivering high-voltage electric pulses that enable the chemotherapeutic drug to more effectively destroy the tumor cells. Electrochemotherapy can be used for treating deep-seated metastases (e.g. in the liver, bone, brain, soft tissue) using variable-geometry long-needle electrodes. To treat deep-seated tumors, patient-specific treatment planning of the electroporation-based treatment is required. Treatment planning is based on generating a 3D model of the organ and target tissue subject to electroporation (i.e. tumor nodules). The generation of the 3D model is done by segmentation algorithms. We implemented and evaluated three automatic liver segmentation algorithms: region growing, adaptive threshold, and active contours (snakes). The algorithms were optimized using a seven-case dataset manually segmented by the radiologist as a training set, and finally validated using an additional four-case dataset that was previously not included in the optimization dataset. The presented results demonstrate that patient's medical images that were not included in the training set can be successfully segmented using our three algorithms. Besides electroporation-based treatments, these algorithms can be used in applications where automatic liver segmentation is required.

Remote access of radiological images using android

Aims: To develop a Graphical User Interface (GUI) system for mobile application to provide secured remote access to medical images for faster diagnostic and therapeutic use. Settings and Design: We have designed an android application using the android operating system. The application is compatible to DICOM images. The application has two layouts and with tools such as window scaling, bone edge detection and RGB colouring. We have also designed a database server for storing patients medical images were a physician can store patient's medical image taken through one of the imaging modality. Methods and Material: The system consists of two separate modules one is at the hospital called Base unit and the other at doctor's site called Mobile unit. At the base unit we have developed a database server for storing the medical images of the patient. We have developed the database using the PHP and my SQL programming languages. The software design and implementation is similar to the client server model. Communication between the two parts is achieved using TCP/IP as network protocol, which ensures safe data transmission over different telecommunication means such as GSM, Satellite, and POTS. The Mobile unit mainly consists of a software module which is built into an android application used for receiving the image data (DICOM Images) of the patient, viewing the image and sending back the report of the patient to the base unit using a communication module (GSM, Satellite or POTS modem). Here the Mobile unit is the mobile phone supporting android applications (smart phones, tablet computers) [5] . An android application viewer is created in the mobile phone of the doctor's through which he can access the medical images stored in the base unit. Results: Proposed work is capable of accessing the medical images from the remote area by the doctor's using their android smart phones. At the mobile unit doctors can use the image enhancement techniques provided for better visualization of the medical images. Also doctor can send the report back to the base unit. Conclusions: As wireless technology is becoming an important part of health care centre's, proposed work intended to use small and portable mobile devices for the assistance of healthcare personnel. It enables the remote access of patient information and images through his wireless handheld devices. Development of such applications will enable to provide better healthcare facilities in time to patients.

Patient-specific treatment planning of electrochemotherapy: Procedure design and possible pitfalls

Electrochemotherapy uses electroporation for enhancing chemotherapy. Electrochemotherapy can be performed using standard operating procedures with predefined electrode geometries, or using patient-specific treatment planning to predict electroporation. The latter relies on realistic computer models to provide optimal results (i.e. electric field distribution as well as electrodes' position and number) and is suitable for treatment of deep-seated tumors.Since treatment planning for deep-seated tumors has been used in radiotherapy, we expose parallelisms with radiotherapy in order to establish the procedure for electrochemotherapy of deep-seated tumors. We partitioned electrochemotherapy in the following phases: the mathematical model of electroporation, treatment planning, set-up verification, treatment delivery and monitoring, and response assessment. We developed a conceptual treatment planning software that incorporates mathematical models of electroporation. Preprocessing and segmentation of the patient's medical images are performed, and a 3D model is constructed which allows placement of electrodes and implementation of the mathematical model of electroporation.We demonstrated the feasibility of electrochemotherapy of deep-seated tumors treatment planning within a clinical study where treatment planning contributed to the effective electrochemotherapy treatment of deep-seated colorectal metastases in the liver. The described procedure can provide medical practitioners with information on using electrochemotherapy in the clinical setting. The main aims of this paper are: 1) to present the procedure for treating deep-seated tumors by electrochemotherapy based on patient-specific treatment planning, and 2) to identify gaps in knowledge and possible pitfalls of such procedure.

Physiome-Model–Based State-Space Framework for Cardiac Deformation Recovery

To more reliably recover cardiac information from noise-corrupted, patient-specific measurements, it is essential to employ meaningful constraining models and adopt appropriate optimization criteria to couple the models with the measurements. Although biomechanical models have been extensively used for myocardial motion recovery with encouraging results, the passive nature of such constraints limits their ability to fully count for the deformation caused by active forces of the myocytes. To overcome such limitations, we propose to adopt a cardiac physiome model as the prior constraint for cardiac motion analysis. The cardiac physiome model comprises an electric wave propagation model, an electromechanical coupling model, and a biomechanical model, which are connected through a cardiac system dynamics for a more complete description of the macroscopic cardiac physiology. Embedded within a multiframe state-space framework, the uncertainties of the model and the patient's measurements are systematically dealt with to arrive at optimal cardiac kinematic estimates and possibly beyond. Experiments have been conducted to compare our proposed cardiac-physiome-model-based framework with the solely biomechanical model-based framework. The results show that our proposed framework recovers more accurate cardiac deformation from synthetic data and obtains more sensible estimates from real magnetic resonance image sequences. With the active components introduced by the cardiac physiome model, cardiac deformations recovered from patient's medical images are more physiologically plausible.

An integrated medical image database and retrieval system using a web application server

We developed an Integrated Medical Image Database and Retrieval System (INIS) for easy access by medical staff. The INIS mainly consisted of four parts: specific servers to save medical images from multi-vendor modalities of CT, MRI, CR, ECG and endoscopy; an integrated image database (DB) server to save various kinds of images in a DICOM format; a Web application server to connect clients to the integrated image DB and the Web browser terminals connected to an HIS system. The INIS provided a common screen design to retrieve CT, MRI, CR, endoscopic and ECG images, and radiological reports, which would allow doctors to retrieve radiological images and corresponding reports, or ECG images of a patient simultaneously on a screen. Doctors working in internal medicine on average accessed information 492 times a month. Doctors working in cardiological and gastroenterological accessed information 308 times a month. Using the INIS, medical staff could browse all or parts of a patient's medical images and reports.