Colon cancer diagnosis by means of explainable deep learning.

Early detection of the adenocarcinoma cancer in colon tissue by means of explainable deep learning, by classifying histological images and providing visual explainability on model prediction. Considering that in recent years, deep learning techniques have emerged as powerful techniques in medical image analysis, offering unprecedented accuracy and efficiency, in this paper we propose a method to automatically detect the presence of cancerous cells in colon tissue images. Various deep learning architectures are considered, with the aim of considering the best one in terms of quantitative and qualitative results. As a matter of fact, we consider qualitative results by taking into account the so-called prediction explainability, by providing a way to highlight on the tissue images the areas that from the model point of view are related to the presence of colon cancer. The experimental analysis, performed on 10,000 colon issue images, showed the effectiveness of the proposed method by obtaining an accuracy equal to 0.99. The experimental analysis shows that the proposed method can be successfully exploited for colon cancer detection and localisation from tissue images.

Explainable Convolutional Neural Networks for Brain Cancer Detection and Localisation.

Brain cancer is widely recognised as one of the most aggressive types of tumors. In fact, approximately 70% of patients diagnosed with this malignant cancer do not survive. In this paper, we propose a method aimed to detect and localise brain cancer, starting from the analysis of magnetic resonance images. The proposed method exploits deep learning, in particular convolutional neural networks and class activation mapping, in order to provide explainability by highlighting the areas of the medical image related to brain cancer (from the model point of view). We evaluate the proposed method with 3000 magnetic resonances using a free available dataset. The results we obtained are encouraging. We reach an accuracy ranging from 97.83% to 99.67% in brain cancer detection by exploiting four different models: VGG16, ResNet50, Alex_Net, and MobileNet, thus showing the effectiveness of the proposed method.

Head-to-head comparison of diagnostic accuracy of stress-only myocardial perfusion imaging with conventional and cadmium-zinc telluride single-photon emission computed tomography in women with suspected coronary artery disease.

BACKGROUND: Breast attenuation may impact the diagnostic accuracy of stress myocardial perfusion imaging (MPI) with single-photon emission computed tomography (SPECT). We compared the performance of conventional (C)-SPECT and cadmium-zinc-telluride (CZT)-SPECT systems in women with low-intermediate likelihood of coronary artery disease (CAD). METHODS AND RESULTS: A total of 109 consecutive women underwent stress-optional rest MPI by both C-SPECT and CZT-SPECT. In the overall study population, a weak albeit significant correlation between total perfusion defect (TPD) measured by C-SPECT and CZT-SPECT was observed (r = 0.38, P < .001) and at Bland-Altman analysis the mean difference in TPD (C-SPECT minus CZT-SPECT) was 2.40% (P < .001). Overall concordance of semi-quantitative diagnostic performance between C-SPECT and CZT-SPECT was observed in 52 (48%) women with a κ value of 0.09. Normalcy rate was significantly higher using CZT-SPECT compared to C-SPECT (P < .001). Machine learning analysis performed through the implementation of J48 algorithm proved that CZT-SPECT has higher sensitivity, specificity, and accuracy than C-SPECT. CONCLUSIONS: In women with low-intermediate likelihood of CAD, there is a poor concordance of diagnostic performance between C-SPECT and CZT-SPECT, and CZT-SPECT allows better normalcy rate detection compared to C-SPECT.

Phenomenological models of NaV1.5. A side by side, procedural, hands-on comparison between Hodgkin-Huxley and kinetic formalisms.

Computational models of ion channels represent the building blocks of conductance-based, biologically inspired models of neurons and neural networks. Ion channels are still widely modelled by means of the formalism developed by the seminal work of Hodgkin and Huxley (HH), although the electrophysiological features of the channels are currently known to be better fitted by means of kinetic Markov-type models. The present study is aimed at showing why simplified Markov-type kinetic models are more suitable for ion channels modelling as compared to HH ones, and how a manual optimization process can be rationally carried out for both. Previously published experimental data of an illustrative ion channel (NaV1.5) are exploited to develop a step by step optimization of the two models in close comparison. A conflicting practical limitation is recognized for the HH model, which only supplies one parameter to model two distinct electrophysiological behaviours. In addition, a step by step procedure is provided to correctly optimize the kinetic Markov-type model. Simplified Markov-type kinetic models are currently the best option to closely approximate the known complexity of the macroscopic currents of ion channels. Their optimization can be achieved through a rationally guided procedure, and allows to obtain models with a computational burden that is comparable with HH models one.

Patient-specific mobility assessment to monitor recovery after total hip arthroplasty.

Total hip arthroplasty is a ubiquitously successful orthopedic surgical procedure, whose prevalence is rising worldwide. While many investigations focus on characterizing periprosthetic pathophysiology, the objective of our research is to develop and describe multi-metric assemblies as a first step toward creating a patient-specific mobility index that rehabilitators and orthopedic surgeons can utilize for prescribing their respective procedures. In total, 48 total hip arthroplasty patients (both cemented and uncemented) undergoing unilateral, primary surgery went through computed tomographic scans and gait analysis measurements both before and 1 year following their surgery. Altogether, the reported quantitative metrics include 11 spatial and temporal gait parameters, muscle density, and electromyography signals from the rectus femoris, vastus lateralis, and vastus medialis, and bone mineral density values from bioimage analysis around the implant stem. We found that measured parameters from a subgroup were sensitive to changes observed during patient recovery, implicating the predictive sensitivity of these patient conditions. Most post-operative gait parameters changed significantly, while electromyography data indicated few significant differences. Moreover, results from bioimage analyses indicate a general reduction of periprosthetic bone mineral density after 1 year, in association with increasing density of the quadriceps muscles. Furthermore, this work identifies which quantitative metrics undergo the greatest variation after total hip arthroplasty and demonstrates the clinical feasibility of a multimodal approach to mobility assessment that may ultimately support decision-making for post-surgical rehabilitation protocols.

Reducing the risk of healthcare-associated infections through Lean Six Sigma: The case of the medicine areas at the Federico II University Hospital in Naples (Italy).

RATIONALE, AIMS, AND OBJECTIVES: Lean Six Sigma (LSS) has been recognized as an effective management tool for improving healthcare performance. Here, LSS was adopted to reduce the risk of healthcare-associated infections (HAIs), a critical quality parameter in the healthcare sector. METHODS: Lean Six Sigma was applied to the areas of clinical medicine (including general medicine, pulmonology, oncology, nephrology, cardiology, neurology, gastroenterology, rheumatology, and diabetology), and data regarding HAIs were collected for 28,000 patients hospitalized between January 2011 and December 2016. Following the LSS define, measure, analyse, improve, and control cycle, the factors influencing the risk of HAI were identified by using typical LSS tools (statistical analyses, brainstorming sessions, and cause-effect diagrams). Finally, corrective measures to prevent HAIs were implemented and monitored for 1 year after implementation. RESULTS: Lean Six Sigma proved to be a useful tool for identifying variables affecting the risk of HAIs and implementing corrective actions to improve the performance of the care process. A reduction in the number of patients colonized by sentinel bacteria was achieved after the improvement phase. CONCLUSIONS: The implementation of an LSS approach could significantly decrease the percentage of patients with HAIs.

Improving performances of the knee replacement surgery process by applying DMAIC principles.

RATIONALE, AIMS, AND OBJECTIVES: The work is a part of a project about the application of the Lean Six Sigma to improve health care processes. A previously published work regarding the hip replacement surgery has shown promising results. Here, we propose an application of the DMAIC (Define, Measure, Analyse, Improve, and Control) cycle to improve quality and reduce costs related to the prosthetic knee replacement surgery by decreasing patients' length of hospital stay (LOS) METHODS: The DMAIC cycle has been adopted to decrease the patients' LOS. The University Hospital "Federico II" of Naples, one of the most important university hospitals in Southern Italy, participated in this study. Data on 148 patients who underwent prosthetic knee replacement between 2010 and 2013 were used. Process mapping, statistical measures, brainstorming activities, and comparative analysis were performed to identify factors influencing LOS and improvement strategies. RESULTS: The study allowed the identification of variables influencing the prolongation of the LOS and the implementation of corrective actions to improve the process of care. The adopted actions reduced the LOS by 42%, from a mean value of 14.2 to 8.3 days (standard deviation also decreased from 5.2 to 2.3 days). CONCLUSIONS: The DMAIC approach has proven to be a helpful strategy ensuring a significant decreasing of the LOS. Furthermore, through its implementation, a significant reduction of the average costs of hospital stay can be achieved. Such a versatile approach could be applied to improve a wide range of health care processes.

Cone-beam micro computed tomography dedicated to the breast.

We developed a scanner for micro computed tomography dedicated to the breast (BµCT) with a high resolution flat-panel detector and a microfocus X-ray tube. We evaluated the system spatial resolution via the 3D modulation transfer function (MTF). In addition to conventional absorption-based X-ray imaging, such a prototype showed capabilities for propagation-based phase-contrast and related edge enhancement effects in 3D imaging. The system limiting spatial resolution is 6.2mm-1 (MTF at 10%) in the vertical direction and 3.8mm-1 in the radial direction, values which compare favorably with the spatial resolution reached by mini focus breast CT scanners of other groups. The BµCT scanner was able to detect both microcalcification clusters and masses in an anthropomorphic breast phantom at a dose comparable to that of two-view mammography. The use of a breast holder is proposed in order to have 1-2min long scan times without breast motion artifacts.

Lean Six Sigma: a new approach to the management of patients undergoing prosthetic hip replacement surgery.

RATIONALE, AIMS AND OBJECTIVES: In 2012, health care spending in Italy reached €114.5 billion, accounting for 7.2% of the Gross Domestic Product (GDP) and 14.2% of total public spending. Therefore, reducing waste in health facilities could generate substantial cost savings. The objective of this study is to show that Lean Six Sigma represents an appropriate methodology for the development of a clinical pathway which allows to improve quality and to reduce costs in prosthetic hip replacement surgery. METHODS: The methodology used for the development of a new clinical pathway was Lean Six Sigma. Problem solving in Lean Six Sigma is the DMAIC (Define, Measure, Analyse, Improve, Control) roadmap, characterized by five operational phases which make possible to reach fixed goals through a rigorous process of defining, measuring, analysing, improving and controlling business problems. RESULTS: The following project indicated several variables influencing the inappropriate prolongation of the length of stay for inpatient treatment and corrective actions were performed to improve the effectiveness and efficiency of the process of care. The average length of stay was reduced from 18.9 to 10.6 days (-44%). CONCLUSION: This article shows there is no trade-off between quality and costs: Lean Six Sigma improves quality and, at the same time, reduces costs.

Robustness to noise of arterial blood flow estimation methods in CT perfusion.

BACKGROUND: Perfusion CT is a technology which allows functional evaluation of tissue vascularity. Due to this potential, it is finding increasing utility in oncology. Although since its introduction continuous advances have interested CT technique, some issues have to be still defined, concerning both clinical and technical aspects. In this study, we dealt with the comparison of two widely employed mathematical models (dual input one compartment model - DOCM - and maximum slope - SM -) analyzing their robustness to the noise. METHODS: We carried out a computer simulation process to quantify effect of noise on the evaluation of an important perfusion parameter (Arterial Blood Flow - BFa) in liver tumours. A total of 4500 liver TAC, corresponding to 3 fixed BFa values, were simulated using different arterial and portal TAC (computed from 5 real CT images) at 10 values of signal to noise ratio (SNR). BFa values were calculated by applying four different algorithms, specifically developed, to these noisy simulated curves. Three algorithms were developed to implement SM (one semiautomatic, one automatic and one automatic with filtering) and the last for the DOCM method. RESULTS: In all the simulations, DOCM provided the best results, i.e., those with the lowest percentage error compared to the reference value of BFa. Concerning SM, the results are variable. Results obtained with the automatic algorithm with filtering are close to the reference value, but only if SNR is higher than 50. Vice versa, results obtained by means of the semiautomatic algorithm gave, in all simulations, the lowest results with the lowest standard deviation of the percentage error. CONCLUSIONS: Since the use of DOCM is limited by the necessity that portal vein is visible in CT scans, significant restriction for patients' follow-up, we concluded that SM can be reliably employed. However, a proper software has to be used and an estimation of SNR would be carried out.

Investigating the role of capacitive coupling between the operating table and the return electrode of an electrosurgery unit in the modification of the current density distribution within the patients' body.

BACKGROUND: Electrosurgery units are widely employed in modern surgery. Advances in technology have enhanced the safety of these devices, nevertheless, accidental burns are still regularly reported. This study focuses on possible causes of sacral burns as complication of the use of electrosurgery. Burns are caused by local densifications of the current, but the actual pathway of current within patient's body is unknown. Numerical electromagnetic analysis can help in understanding the issue. METHODS: To this aim, an accurate heterogeneous model of human body (including seventy-seven different tissues), electrosurgery electrodes, operating table and mattress was build to resemble a typical surgery condition. The patient lays supine on the mattress with the active electrode placed onto the thorax and the return electrode on his back. Common operating frequencies of electrosurgery units were considered. Finite Difference Time Domain electromagnetic analysis was carried out to compute the spatial distribution of current density within the patient's body. A differential analysis by changing the electrical properties of the operating table from a conductor to an insulator was also performed. RESULTS: Results revealed that distributed capacitive coupling between patient body and the conductive operating table offers an alternative path to the electrosurgery current. The patient's anatomy, the positioning and the different electromagnetic properties of tissues promote a densification of the current at the head and sacral region. In particular, high values of current density were located behind the sacral bone and beneath the skin. This did not occur in the case of non-conductive operating table. CONCLUSION: Results of the simulation highlight the role played from capacitive couplings between the return electrode and the conductive operating table. The concentration of current density may result in an undesired rise in temperature, originating burns in body region far from the electrodes. This outcome is concordant with the type of surgery-related sacral burns reported in literature. Such burns cannot be immediately detected after surgery, but appear later and can be confused with bedsores. In addition, the dosimetric analysis suggests that reducing the capacity coupling between the return electrode and the operating table can decrease or avoid this problem.

Assessment of total hip arthroplasty by means of computed tomography 3D models and fracture risk evaluation.

Total hip arthroplasty (THA) can be achieved by using a cemented or noncemented prosthesis. Besides patient's age, weight, and other clinical signs, the evaluation of the quality of the bones is a crucial parameter on which orthopedic surgeons base the choice between cemented and noncemented THA. Although bone density generally decreases with age and a cemented THA is preferred for older subjects, the bone quality of a particular patient should be quantitatively evaluated. This study proposes a new method to quantitatively measure bone density and fracture risk by using 3D models extracted by a preoperative computed tomography (CT) scan of the patient. Also, the anatomical structure and compactness of the quadriceps muscle is computed to provide a more complete view. A spatial reconstruction of the tissues is obtained by means of CT image processing, then a detailed 3D model of bone mineral density of the femur is provided by including quantitative CT density information (CT must be precalibrated). A finite element analysis will provide a map of the strains around the proximal femur socket when solicited by typical stresses caused by an implant. The risk for structural failure due to press-fitting and compressive stress during noncemented THA surgery was estimated by calculating a bone fracture risk index (ratio between actual compressive stress and estimated failure stress of the bone). A clinical trial was carried out including 36 volunteer patients (ages 22-77) who underwent unilateral THA surgery for the first time: 18 received a cemented implant and 18 received a noncemented implant. CT scans were acquired before surgery, immediately after, and after 12 months. Bone and quadriceps density results were higher in the healthy leg in about 80% of the cases. Bone and quadriceps density generally decrease with age but mineral density may vary significantly between patients. Preliminary results indicate the highest fracture risk at the calcar and the lowest at the intertrocanteric line, with some difference between patients. An analysis of the results suggest that this methodology can be a valid noninvasive decision support tool for THA planning; however, further analyses are needed to tune the technique and to allow clinical applications. Combination with gait analysis data is planned.