Research

Boosting multiple sclerosis lesion segmentation through attention mechanism

A. Rondinella, E. Crispino, F. Guarnera, O. Giudice, A. Ortis, G. Russo, C. Di Lorenzo, D. Maimone, F. Pappalardo, S. Battiato. Boosting multiple sclerosis lesion segmentation through attention mechanism. Computers in Biology and Medicine, 2023, 107021.

We proposed a framework that exploits an augmented U-Net architecture with a convolutional long short-term memory layer and attention mechanism which is able to segment and quantify multiple sclerosis lesions detected in magnetic resonance images. Quantitative and qualitative evaluation on challenging examples demonstrated how the method outperforms previous state-of-the-art approaches, reporting an overall Dice score of 89% and also demonstrating robustness and generalization ability on never seen new test samples of a new dedicated under construction dataset.

Attention-Based Convolutional Neural Network for CT Scan COVID-19 Detection

A. Rondinella, F. Guarnera, O. Giudice, A. Ortis, F. Rundo, S. Battiato. Attention-Based Convolutional Neural Network for CT Scan COVID-19 Detection. Proceedings of the IEEE International Conference on Acoustics Speech and Signal Processing Satellite Workshops, IEEE (2023).

We present a solution for Covid-19 detection, presented in the challenge of 3rd Covid-19 competition, inside the "AI-enabled Medical Image Analysis Workshop" organized by IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) 2023. In this work, the application of deep learning models for chest CT image analysis was investigated, focusing on the use of a ResNet as a backbone network augmented with attention mechanisms. The ResNet provides an effective feature extractor for the classification task, while the attention mechanisms improve the model's ability to focus on important regions of interest within the images. We conducted extensive experiments on a provided dataset and achieved a macro F1 score of 0.78 on the test set, demonstrating the potential to assist the diagnosis of Covid-19. Our proposed approach leverages the power of deep learning with attention mechanisms to address the challenges of Covid-19 detection in the early detection and management of the disease. In both test and validation set, the proposed method outperformed the baseline of the challenge, ranking fifth in the competition.

Mixup Data Augmentation for COVID-19 Infection Percentage Estimation

Napoli Spatafora, M. A., Ortis, A., & Battiato, S. (2022, August). Mixup Data Augmentation for COVID-19 Infection Percentage Estimation. In Image Analysis and Processing. ICIAP 2022 Workshops: ICIAP International Workshops, Lecce, Italy, May 23–27, 2022, Revised Selected Papers, Part II (pp. 508-519). Cham: Springer International Publishing.

The outbreak of the COVID-19 pandemic considerably increased the workload in hospitals. In this context, the availability of proper diagnostic tools is very important in the fight against this virus. Scientific research is constantly making its contribution in this direction. Actually, there are many scientific initiatives including challenges that require to develop deep algorithms that analyse X-ray or Computer Tomography (CT) images of lungs. One of these concerns a challenge whose topic is the prediction of the percentage of COVID-19 infection in chest CT images. In this paper, we present our contribution to the COVID-19 Infection Percentage Estimation Competition organised in conjunction with the ICIAP 2021 Conference. The proposed method employs algorithms for classification problems such as Inception-v3 and the technique of data augmentation mixup on COVID-19 images. Moreover, the mixup methodology is applied for the first time in radiological images of lungs affected by COVID-19 infection, with the aim to infer the infection degree with slice-level precision. Our approach achieved promising results despite the specific constrains defined by the rules of the challenge, in which our solution entered in the final ranking.

An Explainable Medical Imaging Framework for Modality Classifications Trained Using Small Datasets

Trenta, F., Battiato, S., & Ravì, D. (2022, May). An Explainable Medical Imaging Framework for Modality Classifications Trained Using Small Datasets. In Image Analysis and Processing–ICIAP 2022: 21st International Conference, Lecce, Italy, May 23–27, 2022, Proceedings, Part I (pp. 358-367). Cham: Springer International Publishing.

With the huge expansion of artificial intelligence in medical imaging, many clinical warehouses, medical centres and research communities, have organized patients’ data in well-structured datasets. These datasets are one of the key elements to train AI-enabled solutions. Additionally, the value of such datasets depends on the quality of the underlying data. To maintain the desired high-quality standard, these datasets are actively cleaned and continuously expanded. This labelling process is time-consuming and requires clinical expertise even when a simple classification task must be performed. Therefore, in this work, we propose to tackle this problem by developing a new pipeline for the modality classification of medical images. Our pipeline has the purpose to provide an initial step in organizing a large collection of data and grouping them by modality, thus reducing the involvement of costly human raters. In our experiments, we consider 4 popular deep neural networks as the core engine of the proposed system. The results show that when limited datasets are available simpler pre-trained networks achieved better results than more complex and sophisticated architectures. We demonstrate this by comparing the considered networks on the ADNI dataset and by exploiting explainable AI techniques that help us to understand our hypothesis. Still today, many medical imaging studies make use of limited datasets, therefore we believe that our contribution is particularly relevant to drive future developments of new medical imaging technologies when limited data are available.

Advanced Deep Network with Attention and Genetic-Driven Reinforcement Learning Layer for an Efficient Cancer Treatment Outcome Prediction

F. Rundo, G. L. Banna, F. Trenta and S. Battiato, "Advanced Deep Network with Attention and Genetic-Driven Reinforcement Learning Layer for an Efficient Cancer Treatment Outcome Prediction," 2021 IEEE International Conference on Image Processing (ICIP), Anchorage, AK, USA, 2021, pp. 294-298, doi: 10.1109/ICIP42928.2021.9506805.

In the last few years, medical researchers have investigated promising approaches for cancer treatment, leading to a major interest in the immunotherapeutic approach. The target of immunotherapy is to boost a subject’s immune system in order to fight cancer. However, scientific studies confirmed that not all patients have a positive response to immunotherapy treatment. Medical research has long been engaged in the search for predictive immunotherapeutic-response biomarkers. Based on these considerations, we developed a non-invasive advanced pipeline with a downstream 3D deep classifier with attention and reinforcement learning for early prediction of patients responsive to immunotherapeutic treatment from related chest-abdomen CT-scan imaging. We have tested the proposed pipeline within a clinical trial that recruited patients with metastatic bladder cancer. Our experiment results achieved accuracy close to 93%.

Advanced Densely Connected System with Embedded Spatio-Temporal Features Augmentation for Immunotherapy Assessment

F. Rundo, G. L. Banna, F. Trenta and S. Battiato, "Advanced Densely Connected System with Embedded Spatio-Temporal Features Augmentation for Immunotherapy Assessment," 2021 International Joint Conference on Neural Networks (IJCNN), Shenzhen, China, 2021, pp. 1-7, doi: 10.1109/IJCNN52387.2021.9534078.

In medical field, the term “immunotherapy” refers to a form of cancer treatment that uses the ability of body's immune system to prevent and destroy cancer cells. In the last few years, immunotherapy has demonstrated to be a very effective treatment in fighting cancer diseases. However, immunotherapy does not work for every patients and moreover, certain types of immunotherapy drugs could have side effects. With this regard, scientific researchers are investigating for effective ways to select the patients who are more likely to respond to the treatment. Hence, pre-clinical data confirmed that, sometimes, the composition of immune system cells infiltrating the tumor micro-environment may interfere with the efficacy of immunotherapy treatments. In this work, we developed a 3D Deep Network with a downstream classifier for selecting and properly augmenting features from chest-abdomen CT images toward improving cancer outcome prediction. In our work, we proposed an effective solution to a specific type of aggressive bladder cancer, called Metastatic Urothelial Carcinoma (mUC). Our experiment results achieved high accuracy confirming the effectiveness of the proposed pipeline.

Interpretable Deep Model For Predicting Gene-Addicted Non-Small-Cell Lung Cancer In Ct Scans

Pino, C., Palazzo, S., Trenta, F., Cordero, F., Bagci, U., Rundo, F., Battiato, S., Giordano, D., Aldinucci, M., & Spampinato, C. (2021, April). Interpretable deep model for predicting gene-addicted non-small-cell lung cancer in ct scans. In 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI) (pp. 891-894). IEEE.

Genetic profiling and characterization of lung cancers have recently emerged as a new technique for targeted therapeutic treatment based on immunotherapy or molecular drugs. However, the most effective way to discover specific gene mutations through tissue biopsy has several limitations, from invasiveness to being a risky procedure. Recently, quantitative assessment of visual features from CT data has been demonstrated to be a valid alternative to biopsy for the diagnosis of gene-addicted tumors. In this paper, we present a deep model for automated lesion segmentation and classification as gene-addicted or not. The segmentation approach extends the 2D Tiramisu architecture for 3D segmentation through dense blocks and squeeze-and-excitation layers, while a multi-scale 3D CNN is used for lesion classification. We also train our model with adversarial samples, and show that this approach acts as a gradient regularizer and enhances model interpretability. We also built a dataset, the first of its nature, consisting of 73 CT scans annotated with the presence of a specific genomics profile. We test our approach on this dataset achieving a segmentation accuracy of 93.11% (Dice score) and a classification accuracy in identifying oncogene-addicted lung tumors of 82.00%.

Advanced 3D Deep Non-Local Embedded System for Self-Augmented X-Ray-Based COVID-19 Assessment

Rundo, F., Genovese, A., Leotta, R., Scotti, F., Piuri, V., & Battiato, S. (2021). Advanced 3d deep non-local embedded system for self-augmented x-ray-based covid-19 assessment. In Proceedings of the IEEE/CVF International Conference on Computer Vision (pp. 423-432).

COVID-19 diagnosis using chest x-ray (CXR) imaging has a greater sensitivity and faster acquisition procedures than the Real-Time Polimerase Chain Reaction (RT-PCR) test, also requiring radiology machinery that is cheap and widely available. To process the CXR images, methods based on Deep Learning (DL) are being increasingly used, often in combination with data augmentation techniques. However, no method in the literature performs data augmentation in which the augmented training samples are processed collectively as a multi-channel image. Furthermore, no approach has yet considered a combination of attention-based networks with Convolutional Neural Networks (CNN) for COVID-19 detection. In this paper, we propose the first method for COVID-19 detection from CXR images that uses an innovative self-augmentation scheme based on reinforcement learning, which combines all the augmented images in a 3D deep volume and processes them together using a novel non-local deep CNN, which integrates convolutional and attention layers based on non-local blocks. Results on publicly-available databases exhibit a greater accuracy than the state of the art, also showing that the regions of CXR images influencing the decision are consistent with radiologists' observations.

Advanced Non-linear Generative Model with a Deep Classifier for Immunotherapy Outcome Prediction: A Bladder Cancer Case Study

Rundo, F., Banna, G. L., Trenta, F., Spampinato, C., Bidaut, L., Ye, X., Kollias, S., & Battiato, S. (2021). Advanced non-linear generative model with a deep classifier for immunotherapy outcome prediction: A bladder cancer case study. In Pattern Recognition. ICPR International Workshops and Challenges: Virtual Event, January 10–15, 2021, Proceedings, Part I (pp. 227-242). Springer International Publishing.

Immunotherapy is one of the most interesting and promising cancer treatments. Encouraging results have confirmed the effectiveness of immunotherapy drugs for treating tumors in terms of long-term survival and a significant reduction in toxicity compared to more traditional chemotherapy approaches. However, the percentage of patients eligible for immunotherapy is rather small, and this is likely related to the limited knowledge of physiological mechanisms by which certain subjects respond to the treatment while others have no benefit. To address this issue, the authors propose an innovative approach based on the use of a non-linear cellular architecture with a deep downstream classifier for selecting and properly augmenting 2D features from chest-abdomen CT images toward improving outcome prediction. The proposed pipeline has been designed to make it usable over an innovative embedded Point of Care system. The authors report a case study of the proposed solution applied to a specific type of aggressive tumor, namely Metastatic Urothelial Carcinoma (mUC). The performance evaluation (overall accuracy close to 93%) confirms the proposed approach effectiveness.

3D Non-Local Neural Network: A Non-Invasive Biomarker for Immunotherapy Treatment Outcome Prediction. Case-Study: Metastatic Urothelial Carcinoma

Rundo, F., Banna, G. L., Prezzavento, L., Trenta, F., Conoci, S., & Battiato, S. (2020). 3d non-local neural network: A non-invasive biomarker for immunotherapy treatment outcome prediction. case-study: Metastatic urothelial carcinoma. Journal of Imaging, 6(12), 133.

Immunotherapy is regarded as one of the most significant breakthroughs in cancer treatment. Unfortunately, only a small percentage of patients respond properly to the treatment. Moreover, to date, there are no efficient bio-markers able to early discriminate the patients eligible for this treatment. In order to help overcome these limitations, an innovative non-invasive deep pipeline, integrating Computed Tomography (CT) imaging, is investigated for the prediction of a response to immunotherapy treatment. We report preliminary results collected as part of a case study in which we validated the implemented method on a clinical dataset of patients affected by Metastatic Urothelial Carcinoma. The proposed pipeline aims to discriminate patients with high chances of response from those with disease progression. Specifically, the authors propose ad-hoc 3D Deep Networks integrating Self-Attention mechanisms in order to estimate the immunotherapy treatment response from CT-scan images and such hemato-chemical data of the patients. The performance evaluation (average accuracy close to 92%) confirms the effectiveness of the proposed approach as an immunotherapy treatment response biomarker.

Breast shape analysis

G. Catanuto, W. Taher, N. Rocco, F. Catalano, D. Allegra, F. Milotta, F. Stanco, G. Gallo, M.B. Nava. Breast Shape Analysis with Curvature Estimates and Principal Component Analysis for Cosmetic and Reconstructive Breast Surgery. Aesthetic Surgery Journal. 2018.

In this work we focus on digital shape analysis of breast models to assist breast surgeon for medical and surgical purposes. A clinical procedure for female breast digital scan is proposed. After a manual ROI definition through cropping, the meshes are automatically processed. The breasts are represented exploiting “bag of normal” representation, resulting in a 64-d descriptor. PCA is computed and the obtained first two principal components are used to plot the breasts shape into a 2D space. We show how the breasts subject to a surgery change their representation in this space and provide a cue about the error in this estimation. We believe that the proposed procedure represents a valid solution to evaluate the results of surgeries, since one of the most important goal of the specialists is to symmetrically reconstruct breasts and an objective tool to measure the result is currently missing.

G. Catanuto, A. Spano, A. Pennati, E. Riggio, G.M. Farinella, G. Impoco, S. Spoto, G. Gallo, M.B. Nava. Experimental methodology for digital breast shape analysis and objective surgical outcome evaluation. Journal of Plastic, Reconstructive & Aesthetic Surgery; 61; 3; 314-318. 2008.

Outcome evaluation in cosmetic and reconstructive surgery of the breast is commonly performed visually or employing bi-dimensional photography. The reconstructive process in the era of anatomical implants requires excellent survey capabilities that mainly rely on surgeon experience. In this paper we present a set of parameters to unambiguously estimate the shape of natural and reconstructed breast. A digital laser scanner was employed on seven female volunteers. A graphic depiction of curvature of the thoracic surface has been the most interesting result.

Objective analysis of simple kidney cysts from CT images

S. Battiato, G.M. Farinella, G. Gallo, O. Garretto, C. Privitera. Objective analysis of simple kidney cysts from CT images. Medical Measurements and Applications (MeMeA), IEEE International Workshop, 146-149. 2009.

Simple kidney cysts analysis from CT images is nowadays performed in a direct visual and hardly reproducible way. Computer-aided measurements of simple kidney cysts from CT images may help radiologists to accomplish an objective analysis of the clinical cases under observation. We proposed a semi-automatic segmentation algorithm for this task. Experiments performed on real datasets confirm the effectiveness and usefulness of the proposed method.

Neurofuzzy Segmentation of Microarray Images

S. Battiato, G.M. Farinella, G. Gallo, G. C. Guarnera. Neurofuzzy Segmentation of Microarray Images. Proceedings of IAPR International Conference on Pattern Recognition (ICPR), pp. 1-4. Tampa, Florida, USA, 2008.

We proposed a novel microarray segmentation strategy to separate background and foreground signals in microarray images making use of a neurofuzzy processing pipeline. In particular a Kohonen Self Organizing Map followed by a Fuzzy K-Mean classifier are employed to properly manage critical cases like saturated spot and spike noise. To speed up the overall process a Hilbert sampling is performed together with an ad-hoc analysis of statistical distribution of signals. Experiments confirm the validity of the proposed technique both in terms of measured and visual inspection quality.