Intelligent system based on multiple networks for accurate ovarian tumor semantic segmentation.

Ovarian tumors, especially malignant ones, represent a global concern, with increased prevalence in recent years. More accurate medical support systems are urgently needed to support medical staff in obtaining an efficient ovarian tumors diagnosis since detection in early stages could lead to immediately applying appropriate treatment, and implicitly improving the survival rate. The current paper aims to demonstrate that more accurate systems could be designed by combining different convolutional neural networks using different custom combination approaches and by selecting the appropriate networks to be involved in the ensemble model to achieve the best performance metrics. It is essential to understand if combining all experimented networks or only the best-performing ones could always lead to the most effective results or not. The current paper is structured in three main phases. The first step is to propose the individual networks involved in the experiments. Five DeepLab-V3+ networks with different encoders (ResNet-18, ResNet-50, MobileNet-V2, InceptionResNet-V2, and Xception) were used. In the second step, the paper proposes a custom algorithm to combine multiple individual semantic segmentation networks, while the last step describes the iterative selection approach for selecting all individual networks to be combined so that the most accurate ensemble is obtained. The system performing semantic segmentation for different types of ovarian tumors, covering both benign and malignant ones, achieved 91.18 % Intersection over union (IoU), thus overperforming all individual networks. The proposed method could be extended so that more powerful deep learning models could be used.

Detection of Malignant Skin Lesions Based on Decision Fusion of Ensembles of Neural Networks.

Today, skin cancer, and especially melanoma, is an increasing and dangerous health disease. The high mortality rate of some types of skin cancers needs to be detected in the early stages and treated urgently. The use of neural network ensembles for the detection of objects of interest in images has gained more and more interest due to the increased performance of the results. In this sense, this paper proposes two ensembles of neural networks, based on the fusion of the decisions of the component neural networks for the detection of four skin lesions (basal cancer cell, melanoma, benign keratosis, and melanocytic nevi). The first system is based on separate learning of three neural networks (MobileNet V2, DenseNet 169, and EfficientNet B2), with multiple weights for the four classes of lesions and weighted overall prediction. The second system is made up of six binary models (one for each pair of classes) for each network; the fusion and prediction are conducted by weighted summation per class and per model. In total, 18 such binary models will be considered. The 91.04% global accuracy of this set of binary models is superior to the first system (89.62%). Separately, only for the binary classifications within the system was the individual accuracy better. The individual F1 score for each class and the global system varied from 81.36% to 94.17%. Finally, a critical comparison is made with similar works from the literature.

Decision Support System for Liver Lesion Segmentation Based on Advanced Convolutional Neural Network Architectures.

Given its essential role in body functions, liver cancer is the third most common cause of death from cancer, despite being the sixth most common type of cancer worldwide. Following advancements in medicine and image processing, medical image segmentation methods are receiving a great deal of attention. As a novelty, the paper proposes an intelligent decision system for segmenting liver and hepatic tumors by integrating four efficient neural networks (ResNet152, ResNeXt101, DenseNet201, and InceptionV3). Images from computed tomography for training, validation, and testing were taken from the public LiTS17 database and preprocessed to better highlight liver tissue and tumors. Global segmentation is done by separately training individual classifiers and the global system of merging individual decisions. For the aforementioned application, classification neural networks have been modified for semantic segmentation. After segmentation based on the neural network system, the images were postprocessed to eliminate artifacts. The segmentation results obtained by the system were better, from the point of view of the Dice coefficient, than those obtained by the individual networks, and comparable with those reported in recent works.

Quality-of-Life Assessment after Head and Neck Oncological Surgery for Advanced-Stage Tumours.

Squamous cell carcinoma of the head and neck (HNSCC) is a common malignancy often diagnosed in the advanced stage with a complex negative influence on the patient's quality of life (QoL). Given its multi-modal treatment, the first step is to adequately balance the needs of the patient, and the second step includes the consultations, interventions, and care provided by the medical team, with the purpose of improving the overall management of the HNSCC. Current attempts to develop and validate quality-of-life instruments specific to cancers of the head and neck have been reported, and certain questionnaires are now available. We performed a retrospective study in a tertiary centre, involving 89 patients who survived 3 years after HNSCC surgery. A patient-related outcome measurement was made using the European Organization for Research and Treatment of Cancer (EORTC) QLQ-C30 and QLQ-H&N35 instruments to assess QoL at admission and 3 years after treatment. The 3-year survivors reported an overall improvement in QoL compared with those in the pre-treatment period. The unique details of head and neck cancer treatments outline the importance of considering the characteristics of the patient population in quality-of-life research and also identify how quality-of-life data can contribute to the care provided by the multi-disciplinary team involved in a patient's follow-up.

Skin Lesion Classification Using Collective Intelligence of Multiple Neural Networks.

Skin lesion detection and analysis are very important because skin cancer must be found in its early stages and treated immediately. Once installed in the body, skin cancer can easily spread to other body parts. Early detection would represent a very important aspect since, by ensuring correct treatment, it could be curable. Thus, by taking all these issues into consideration, there is a need for highly accurate computer-aided systems to assist medical staff in the early detection of malignant skin lesions. In this paper, we propose a skin lesion classification system based on deep learning techniques and collective intelligence, which involves multiple convolutional neural networks, trained on the HAM10000 dataset, which is able to predict seven skin lesions including melanoma. The convolutional neural networks experimentally chosen, considering their performances, to implement the collective intelligence-based system for this purpose are: AlexNet, GoogLeNet, GoogLeNet-Places365, MobileNet-V2, Xception, ResNet-50, ResNet-101, InceptionResNet-V2 and DenseNet201. We then analyzed the performances of each of the above-mentioned convolutional neural networks to obtain a weight matrix whose elements are weights associated with neural networks and classes of lesions. Based on this matrix, a new decision matrix was used to build the multi-network ensemble system (Collective Intelligence-based System), combining each of individual neural network decision into a decision fusion module (Collective Decision Block). This module would then have the responsibility to take a final and more accurate decision related to the prediction based on the associated weights of each network output. The validation accuracy of the proposed system is about 3 percent better than that of the best performing individual network.

New Trends in Melanoma Detection Using Neural Networks: A Systematic Review.

Due to its increasing incidence, skin cancer, and especially melanoma, is a serious health disease today. The high mortality rate associated with melanoma makes it necessary to detect the early stages to be treated urgently and properly. This is the reason why many researchers in this domain wanted to obtain accurate computer-aided diagnosis systems to assist in the early detection and diagnosis of such diseases. The paper presents a systematic review of recent advances in an area of increased interest for cancer prediction, with a focus on a comparative perspective of melanoma detection using artificial intelligence, especially neural network-based systems. Such structures can be considered intelligent support systems for dermatologists. Theoretical and applied contributions were investigated in the new development trends of multiple neural network architecture, based on decision fusion. The most representative articles covering the area of melanoma detection based on neural networks, published in journals and impact conferences, were investigated between 2015 and 2021, focusing on the interval 2018-2021 as new trends. Additionally presented are the main databases and trends in their use in teaching neural networks to detect melanomas. Finally, a research agenda was highlighted to advance the field towards the new trends.

Predicting risk of sudden cardiac death in patients with cardiac sarcoidosis using multimodality imaging and personalized heart modeling in a multivariable classifier.

Cardiac sarcoidosis (CS), an inflammatory disease characterized by formation of granulomas in the heart, is associated with high risk of sudden cardiac death (SCD) from ventricular arrhythmias. Current "one-size-fits-all" guidelines for SCD risk assessment in CS result in insufficient appropriate primary prevention. Here, we present a two-step precision risk prediction technology for patients with CS. First, a patient's arrhythmogenic propensity arising from heterogeneous CS-induced ventricular remodeling is assessed using a novel personalized magnetic-resonance imaging and positron-emission tomography fusion mechanistic model. The resulting simulations of arrhythmogenesis are fed, together with a set of imaging and clinical biomarkers, into a supervised classifier. In a retrospective study of 45 patients, the technology achieved testing results of 60% sensitivity [95% confidence interval (CI): 57-63%], 72% specificity [95% CI: 70-74%], and 0.754 area under the receiver operating characteristic curve [95% CI: 0.710-0.797]. It outperformed clinical metrics, highlighting its potential to transform CS risk stratification.

Deep Learning-Based Methods for Automatic Diagnosis of Skin Lesions.

The main purpose of the study was to develop a high accuracy system able to diagnose skin lesions using deep learning-based methods. We propose a new decision system based on multiple classifiers like neural networks and feature-based methods. Each classifier (method) gives the final decision system a certain weight, depending on the calculated accuracy, helping the system make a better decision. First, we created a neural network (NN) that can differentiate melanoma from benign nevus. The NN architecture is analyzed by evaluating it during the training process. Some biostatistic parameters, such as accuracy, specificity, sensitivity, and Dice coefficient are calculated. Then, we developed three other methods based on convolutional neural networks (CNNs). The CNNs were pre-trained using large ImageNet and Places365 databases. GoogleNet, ResNet-101, and NasNet-Large, were used in the enumeration order. CNN architectures were fine-tuned in order to distinguish the different types of skin lesions using transfer learning. The accuracies of the classifications were determined. The last proposed method uses the classical method of image object detection, more precisely, the one in which some features are extracted from the images, followed by the classification step. In this case, the classification was done by using a support vector machine. Just as in the first method, the sensitivity, specificity, Dice similarity coefficient and accuracy are determined. A comparison of the obtained results from all the methods is then done. As mentioned above, the novelty of this paper is the integration of these methods in a global fusion-based decision system that uses the results obtained by each individual method to establish the fusion weights. The results obtained by carrying out the experiments on two different free databases shows that the proposed system offers higher accuracy results.

Proximal tibial osteosarcoma in young patients: early diagnosis, modular reconstruction.

Osteosarcoma is the most common bone tumor that occurs in children and young adults with prevalence of teenage. There can be identified many subtypes of osteosarcoma by how they look on X-rays and under the microscope. Osteosarcoma can be classified as high-grade, intermediate grade, or low-grade. This has a significant prognostic value of tumor development suggesting the growth rate and the potential for expansion. Between 2009-2013, in the Department of Orthopedics and Traumatology, University Emergency Hospital of Bucharest, Romania, were treated seven cases of osteosarcoma of the proximal third of the tibia in young, early-diagnosed cases without metastasis. The treatment involved resection of tumor formation and reconstruction with a modular prosthesis. Postoperative patients were mobilized for a week without charging the operated limb under the protection of orthesis. During this period continued active and passive mobilization of the ankle and foot to prevent stiffness and to reduce postoperative swelling. From the second postoperative week, patients are mobilizing with progressive charging but not being allowed to do any flexion in order to protect de insertion of medial gastrocnemius muscle rotation flap used to cover the prosthesis and to protect the patellar tendon reinsertion. This extensive surgery does not improve survival rate of these patients compared to treatment by amputation of this pathology but greatly increases the comfort of life and in all cases ensure socio-professional reintegration of these patients. To ensure optimal postoperative results perform a complete diagnosis and preoperative oncological treatment before surgery, if applicable.

A rare case of acute lymphoblastic leukemia in a patient with light chain (AL) amyloidosis treated with lenalidomide.

Lenalidomide belongs to a novel class of drugs called Immunomodulators which are now being used for the treatment of plasma cell dyscrasias with variable degrees of efficacy and toxicity. Though Second Primary Malignancies (SPM) have been a concern with its use, the benefits of the treatment outweigh the risks. The leukemogenic risk seems to be potentiated especially when combined with alkylating agents and the SPMs documented are predominantly myeloblastic. To date there are no reported cases of new lymphocytic leukemias in AL amyloidosis, regardless of whether undergone treatment or not. We present a case of AL amylodosis who was treated with lenalidomide and subsequently developed acute lymphoblastic leukemia.

Transformation of the 5q- syndrome to acute lymphoblastic leukemia: a report of two cases and review of the literature.

Myelodysplastic syndrome (MDS) with an isolated deletion of the long arm of chromosome 5 (5q- syndrome) is a distinct subtype of MDS with an indolent course that rarely transforms to acute leukemia. Deletion of the long arm of chromosome 5 has also been reported in rare cases of de novo B-lymphoblastic leukemia. We present two cases of 5q- syndrome with a similar and unusual course of transformation to lymphoblastic leukemia while on Lenalidomide. These two patients achieved an initial response; however, later acquired a second cytogenetic abnormality, became refractory to treatment and evolved into acute leukemia. At the time of transformation, both patients had recurrence of the 5q- abnormality. Review of the literature and the mechanisms of transformation of the 5q-syndrome into an acute leukemia are discussed. Although the relationship between the events in our cases remains unclear, the intriguing similarity between the two cases raises a question whether immune modulators can alter the natural course of MDS. To our knowledge, no similar cases were previously reported in the literature.

Topology modification for surgical simulation using precomputed finite element models based on linear elasticity.

Surgical simulators provide another tool for training and practising surgical procedures, usually restricted to the use of cadavers. Our surgical simulator utilises Finite Element (FE) models based on linear elasticity. It is driven by displacements, as opposed to forces, allowing for realistic simulation of both deformation and haptic response at real-time rates. To achieve demanding computational requirements, the stiffness matrix K, which encompasses the geometrical and physical properties of the object, is precomputed, along with K⁻¹. Common to many surgical procedures is the requirement of cutting tissue. Introducing topology modifications, such as cutting, into these precomputed schemes does however come as a challenge, as the precomputed data needs to be modified, to reflect the new topology. In particular, recomputing K⁻¹ is too costly to be performed during the simulation. Our topology modification method is based upon updating K⁻¹ rather than entirely recomputing the matrix. By integrating condensation, we improve efficiency to allow for interaction with larger models. We can further enhance this by redistributing computational load to improve the system's real-time response. We exemplify our techniques with results from our surgical simulation system.

Progressive update approach to real-time cutting of finite element models in surgical simulation.

We present an extension of our work on topology modification and deformation for Finite Element Models, in which the inverse stiffness matrix is updated rather than recomputed entirely. Previously we integrated condensation to allow for realistic interaction with larger models. We improve on this by redistributing computational load to increase the system's real-time response. Removing a tetrahedron only requires data associated with the nodes of that tetrahedron, and the surface nodes, to be updated, in order to drive the simulation. However, the update procedure itself needs the entire data structure to be updated. The equations used to update the inverse stiffness matrix are split up such that calculations are only performed for the affected nodes. Data regions corresponding to the surface nodes necessary for deformation calculations are computed immediately, whilst remaining regions can be computed as required, resulting in up to a ten-fold improvement in system response times.

Efficient topology modification and deformation for finite element models using condensation.

Cuts on deformable organs are a central task in the set of physical operations needed in many surgical simulation environments. Our BioMedIA surgical simulator structures are Finite Element Models, driven by displacements on the touched nodes as opposed to forces. This approach allows for realistic simulation of both deformation and haptic response at real-time rates. The integration of condensation into our system increases its efficiency, and allows for complex interaction on larger meshes than would normally be permitted with available memory. We present an extension of our novel algorithm for cuts in deformable organs, in the context of condensed matrices. We show results from our surgical simulation system with real-time haptic feedback.

Multiple contact approach to collision modelling in surgical simulation.

In this paper we present a technique for the modelling of realistic collisions between arbitrary rigid surgical tools and deformable geometry that is independent of the resolution of colliding objects. We use a spatial hash table to provide an efficient narrow-phase collision detection and modelling backend. This is combined with previous work on collision modelling in our surgical simulation environment to model realistic collisions and collision response at haptic rates.