Geochemistry and 14C dating of guano deposits in the Karaftu Cave, Kurdistan, Iran: implication for paleoenvironment.

Karaftu Cave in the northwest of Divandareh includes four floors plus an underground area. The bat hall and its underground area are covered with guano deposits. 14C dating indicates the onset of guano deposition is about 14,260±50 BP years, and its average accumulation rate is about 4.1 mm/yr for depths of 360 to 205 cm. Bacterial and fungal metabolisms decay guano, release acids, and disperse large amounts of microorganisms inside the cave. Interactions between acids and guano caused leaching, dissolution, change in the distribution, and abundance of elements, which leads to the formation of secondary minerals in guano. These variations in minerals and elements also depend on the local climatic conditions. Distribution of elements in the Ce/Ce* versus Pr/Pr* diagram and the correlation coefficient between Ce and Mn display three different paleoclimate conditions (dry, wet, and dry) during the accumulation of the guano. Also, dolomite, phosphate, and iron oxide minerals have been formed during the passage of water through bedrock beneath the guano. This water is unsuitable for drinking and harmful to the organisms in the region. Effect of these acids on the substrate also leads to the collapse of the cave floor, generation of a new underground, fall in the groundwater level, change in the groundwater drainage system, drying of springs around the cave, loss of green cover, and a negative impact on the ecosystem in the region. To keep the cave environment clean and remove these problems, it is better to harvest guano and use it as fertilizer.

A soy lecithin nanoparticles-based extender effectively cryopreserves Holstein bull sperm.

Plant-based semen extenders, typically derived from soybean lecithin, are easier to modulate more and consistent in their composition than animal-based extenders. As large lecithin particles can, however, reduce effectiveness and solubility in bull semen extenders, sonication was used to create nano-lecithin (NL) particles of soybean lecithin. The objective was to determine the effects of lecithin type and concentration on the quality of frozen-thawed bovine sperm. We hypothesized that reducing the size of lecithin improves its interactions with the sperm and enhances the parameters that favor its motility, viability and fertility. Semen was collected from six mature Holstein bulls and ejaculates meeting minimum standards were pooled. Eight Tris-based extenders that contained 1, 2, 3, or 4 % of either conventional lecithin (L1-L4) or NL (NL1-NL4), plus two control extenders (one animal-based extender containing 20 % egg yolk [EY] and a commercial lecithin-based extender [BioXcell®]) were compared. Among soybean lecithin-based extenders, NL3 had the highest total and progressive sperm motility, and average path, straight-line and curvilinear sperm velocity, and was comparable to EY. Additionally, sperm mitochondrial activity was the highest in NL3, whereas sperm viability was highest in EY, NL3, and L4. Following in vitro fertilization of in vitro-matured bovine oocyes, NL3 had cleavage and hatching rates comparable to BioXcell®, but a lower blastocyst rate than EY. Overall, NL3 performed better than the other extenders for most end points, with efficiency comparable to EY. We, therefore, concluded that reducing lecithin particle size to a nano level improves sperm cryopreservation with optimal performance with 3 % NL.

Gene expression profile of placentomes and clinical parameters in the cows with retained placenta.

BACKGROUND: Retained placenta (RP) is a prevalent disorder in cattle with many health-related and economic costs for the farm owners. Its etiology has not been clarified yet and there is no definite therapy for this disorder. In this study we conducted RNA-seq, hematologic and histologic experiments to survey the causes of RP development. METHODS: Blood samples were collected from 4 RP and 3 healthy cows during periparturtion period for hematological assessments followed by placentome sampling within 30 min after parturition. Cows were grouped as RP and control in case the placenta was retained or otherwise expelled, respectively. Total RNA was extracted from placentome samples followed by RNA-sequencing. RESULTS: We showed 240 differentially expressed genes (DEGs) between the RP and control groups. Enrichment analyzes indicated immune system and lipid metabolism as prominent over- and under-represented pathways in RP cows, respectively. Hormonal assessments showed that estradiol-17ß (E2) was lower and cortisol tended to be higher in RP cows compared to controls at the day of parturition. Furthermore, histologic experiment showed that villi-crypt junctions remain tighter in RP cows compared to controls and the crypts layer seemed thicker in the placentome of RP cows. Complete blood cell (CBC) parameters were not significantly different between the two groups. CONCLUSION: Overall, DEGs derived from expression profiling and these genes contributed to enrichment of immune and lipid metabolism pathways. We suggested that E2 could be involved in development of RP and the concentrations of P4 and CBC counts periparturition might not be a determining factor.

Transient ischemic dilation or transient RV visualization in patients with normal SPECT stress myocardial perfusion imaging: Correlation with CT coronary artery calcium scoring and coronary angiography.

BACKGROUND: Ancillary findings on MPI, such as transient ischemic dilation (TID) and transient right ventricular visualization (TRV), are recognized as markers of extensive CAD and predictive of adverse outcomes. They usually occur in association with stress-induced regional MPI abnormalities. However, the clinical significance of these ancillary markers in the presence of normal stress MPI is incompletely understood. METHODS: From a cohort of 564 consecutive patients referred for clinical SPECT stress MPI, 44 patients had normal stress SPECT MPI and either TID (n = 28) or TRV (n = 16). These imaging findings were correlated with CT coronary calcium (CAC), CT coronary angiography (CTA), and invasive coronary angiography (ICA) in patients with severe CAC ≥ 1000 HU. TID and TRV were quantified as stress/rest ratios. Severe CAD was defined as > 70% luminal stenosis on CTA or ICA. RESULTS: The median TID ratio was 1.23, with a range of 1.13-1.48; the median TRV ratio was 1.30, with a range of 1.20-1.48. Of 44 patients with TID or TRV, only 9 patients (20.5%) had severe obstructive > 70% CAD by angiography (6 of 28 patients (21.5%) with TID and 3 of 16 patients (19%) with TRV). Severe multi-vessel CAD occurred in only 2 of 44 patients (4.5%). In contrast, of 9 patients with CAC > 1000 HU, 6 (67%) had severe obstructive CAD. CONCLUSION: In patients with normal stress SPECT MPI and TID or TRV, the incidence of severe obstructive CAD was relatively low and predominantly single-vessel CAD. These findings do not support the concept that TID or TRV with normal stress MPI is predictive of high-risk CAD.

Towards Automatic Prediction of Outcome in Treatment of Cerebral Aneurysms.

Intrasaccular flow disruptors treat cerebral aneurysms by diverting the blood flow from the aneurysm sac. Residual flow into the sac after the intervention is a failure that could be due to the use of an undersized device, or to vascular anatomy and clinical condition of the patient. We report a machine learning model based on over 100 clinical and imaging features that predict the outcome of wide-neck bifurcation aneurysm treatment with an intrasaccular embolization device. We combine clinical features with a diverse set of common and novel imaging measurements within a random forest model. We also develop neural network segmentation algorithms in 2D and 3D to contour the sac in angiographic images and automatically calculate the imaging features. These deliver 90% overlap with manual contouring in 2D and 83% in 3D. Our predictive model classifies complete vs. partial occlusion outcomes with an accuracy of 75.31%, and weighted F1-score of 0.74.

Creation and validation of a chest X-ray dataset with eye-tracking and report dictation for AI development.

We developed a rich dataset of Chest X-Ray (CXR) images to assist investigators in artificial intelligence. The data were collected using an eye-tracking system while a radiologist reviewed and reported on 1,083 CXR images. The dataset contains the following aligned data: CXR image, transcribed radiology report text, radiologist's dictation audio and eye gaze coordinates data. We hope this dataset can contribute to various areas of research particularly towards explainable and multimodal deep learning/machine learning methods. Furthermore, investigators in disease classification and localization, automated radiology report generation, and human-machine interaction can benefit from these data. We report deep learning experiments that utilize the attention maps produced by the eye gaze dataset to show the potential utility of this dataset.

A deep community based approach for large scale content based X-ray image retrieval.

A computer assisted system for automatic retrieval of medical images with similar image contents can serve as an efficient management tool for handling and mining large scale data, and can also be used as a tool in clinical decision support systems. In this paper, we propose a deep community based automated medical image retrieval framework for extracting similar images from a large scale X-ray database. The framework integrates a deep learning-based image feature generation approach and a network community detection technique to extract similar images. When compared with the state-of-the-art medical image retrieval techniques, the proposed approach demonstrated improved performance. We evaluated the performance of the proposed method on two large scale chest X-ray datasets, where given a query image, the proposed approach was able to extract images with similar disease labels with a precision of 85%. To the best of our knowledge, this is the first deep community based image retrieval application on large scale chest X-ray database.

Comparison of Chest Radiograph Interpretations by Artificial Intelligence Algorithm vs Radiology Residents.

Importance: Chest radiography is the most common diagnostic imaging examination performed in emergency departments (EDs). Augmenting clinicians with automated preliminary read assistants could help expedite their workflows, improve accuracy, and reduce the cost of care. Objective: To assess the performance of artificial intelligence (AI) algorithms in realistic radiology workflows by performing an objective comparative evaluation of the preliminary reads of anteroposterior (AP) frontal chest radiographs performed by an AI algorithm and radiology residents. Design, Setting, and Participants: This diagnostic study included a set of 72 findings assembled by clinical experts to constitute a full-fledged preliminary read of AP frontal chest radiographs. A novel deep learning architecture was designed for an AI algorithm to estimate the findings per image. The AI algorithm was trained using a multihospital training data set of 342 126 frontal chest radiographs captured in ED and urgent care settings. The training data were labeled from their associated reports. Image-based F1 score was chosen to optimize the operating point on the receiver operating characteristics (ROC) curve so as to minimize the number of missed findings and overcalls per image read. The performance of the model was compared with that of 5 radiology residents recruited from multiple institutions in the US in an objective study in which a separate data set of 1998 AP frontal chest radiographs was drawn from a hospital source representative of realistic preliminary reads in inpatient and ED settings. A triple consensus with adjudication process was used to derive the ground truth labels for the study data set. The performance of AI algorithm and radiology residents was assessed by comparing their reads with ground truth findings. All studies were conducted through a web-based clinical study application system. The triple consensus data set was collected between February and October 2018. The comparison study was preformed between January and October 2019. Data were analyzed from October to February 2020. After the first round of reviews, further analysis of the data was performed from March to July 2020. Main Outcomes and Measures: The learning performance of the AI algorithm was judged using the conventional ROC curve and the area under the curve (AUC) during training and field testing on the study data set. For the AI algorithm and radiology residents, the individual finding label performance was measured using the conventional measures of label-based sensitivity, specificity, and positive predictive value (PPV). In addition, the agreement with the ground truth on the assignment of findings to images was measured using the pooled κ statistic. The preliminary read performance was recorded for AI algorithm and radiology residents using new measures of mean image-based sensitivity, specificity, and PPV designed for recording the fraction of misses and overcalls on a per image basis. The 1-sided analysis of variance test was used to compare the means of each group (AI algorithm vs radiology residents) using the F distribution, and the null hypothesis was that the groups would have similar means. Results: The trained AI algorithm achieved a mean AUC across labels of 0.807 (weighted mean AUC, 0.841) after training. On the study data set, which had a different prevalence distribution, the mean AUC achieved was 0.772 (weighted mean AUC, 0.865). The interrater agreement with ground truth finding labels for AI algorithm predictions had pooled κ value of 0.544, and the pooled κ for radiology residents was 0.585. For the preliminary read performance, the analysis of variance test was used to compare the distributions of AI algorithm and radiology residents' mean image-based sensitivity, PPV, and specificity. The mean image-based sensitivity for AI algorithm was 0.716 (95% CI, 0.704-0.729) and for radiology residents was 0.720 (95% CI, 0.709-0.732) (P = .66), while the PPV was 0.730 (95% CI, 0.718-0.742) for the AI algorithm and 0.682 (95% CI, 0.670-0.694) for the radiology residents (P < .001), and specificity was 0.980 (95% CI, 0.980-0.981) for the AI algorithm and 0.973 (95% CI, 0.971-0.974) for the radiology residents (P < .001). Conclusions and Relevance: These findings suggest that it is possible to build AI algorithms that reach and exceed the mean level of performance of third-year radiology residents for full-fledged preliminary read of AP frontal chest radiographs. This diagnostic study also found that while the more complex findings would still benefit from expert overreads, the performance of AI algorithms was associated with the amount of data available for training rather than the level of difficulty of interpretation of the finding. Integrating such AI systems in radiology workflows for preliminary interpretations has the potential to expedite existing radiology workflows and address resource scarcity while improving overall accuracy and reducing the cost of care.

Combining Deep Learning and Knowledge-driven Reasoning for Chest X-Ray Findings Detection.

The application of deep learning algorithms in medical imaging analysis is a steadily growing research area. While deep learning methods are thriving in the medical domain, they seldom utilize the rich knowledge associated with connected radiology reports. The knowledge derived from these reports can be utilized to enhance the performance of deep learning models. In this work, we used a comprehensive chest X-ray findings vocabulary to automatically annotate an extensive collection of chest X-rays using associated radiology reports and a vocabulary-driven concept annotation algorithm. The annotated X-rays are used to train a deep neural network classifier for finding detection. Finally, we developed a knowledge-driven reasoning algorithm that leverages knowledge learned from X-ray reports to improve upon the deep learning module's performance on finding detection. Our results suggest that combining deep learning and knowledge from radiology reports in a hybrid framework can significantly enhance overall performance in the CXR finding detection.

AI Accelerated Human-in-the-loop Structuring of Radiology Reports.

Rule-based Natural Language Processing (NLP) pipelines depend on robust domain knowledge. Given the long tail of important terminology in radiology reports, it is not uncommon for standard approaches to miss items critical for understanding the image. AI techniques can accelerate the concept expansion and phrasal grouping tasks to efficiently create a domain specific lexicon ontology for structuring reports. Using Chest X-ray (CXR) reports as an example, we demonstrate that with robust vocabulary, even a simple NLP pipeline can extract 83 directly mentioned abnormalities (Ave. recall=93.83%, precision=94.87%) and 47 abnormality/normality descriptions of key anatomies. The richer vocabulary enables identification of additional label mentions in 10 out of 13 labels (compared to baseline methods). Furthermore, it captures expert insight into critical differences between observed and inferred descriptions, and image quality issues in reports. Finally, we show how the CXR ontology can be used to anatomically structure labeled output.

Genetic association study of promoter variation rs3761549 in the FOXP3 gene of Iranian patients diagnosed with brain tumour.

Forkhead box P3 (FOXP3) gene (Gene ID: 50943, Xp11.23) is an X-linked gene that encodes FOXP3 protein, an essential transcription factor in CD4+ CD25+ FOXP3 regulatory T (Treg) cells. FOXP3 mutation has been linked with the pathogenesis of several tumours; however, little is known about the role of single-nucleotide polymorphism (SNP) in its promoter region and its correlation with brain tumour. In the present study, we have investigated the association between SNPs in the promoter region of FOXP3 gene, a promoter SNP, -2383 C/T (rs3761549) with susceptibility to brain cancer in a population of Iran. The distribution of case, control, age and sex was balanced and with rs3761549 C/T allele frequencies distribution also falling in Hardy-Weinberg equilibrium (P = 0.053 and 0.062). The allele C of rs3761549 is lower in the brain tumour cases when compared with the controls (364 vs 392, P = 0.005). The frequency of combined T variant genotype (TT + CT) was significantly higher in the brain cancer cases compared with the controls (28 vs 8, P = 0.001), which was consistent with the T allele distribution. When we used the CC genotype as a reference, we found that both CT and TT genotypes were associated with a higher risk of developing brain tumour (odds ratio [OR], 0.3583; 95% confidence interval [CI], 0.164-0.8197 and OR, 0; 95% CI, 0-0.4118, respectively).

Comparison of energy consumption of wheat production in conservation and conventional agriculture using DEA.

Energy is one of the essential resources for human life and mainly classified as non-renewable resources. Since huge amounts of energy are consumed in the agriculture sector, an energy audit is an essential strategy in countries. Conservation agriculture as a tool for sustainable development can lead to saving agricultural resources. In the current investigation, energy audit for wheat conservation and conventional production systems was performed. For this purpose, 48 farms were selected randomly in 2016, and their energy performance was evaluated and compared. The data were analyzed to calculate energy parameters. Also, data envelopment analysis technique was used to identify the possible ways to achieve higher efficiency in farms. To this end, current and optimum situations and saving energy in different cultivation systems were determined using Charnes, Cooper, and Rhodes (CCR) model. The research results showed that the average energy ratio, net energy gain, specific energy, and energy productivity for conservation farms were 4.31, 137,656 MJ ha-1, 5.56 MJ kg-1, and 0.18 kg MJ-1, respectively. Corresponded values for conventional farms were measured to be 3.03, 90,101 MJ ha-1, 7.69 MJ kg-1, and 0.13 kg MJ-1, respectively. Data envelopment analysis results revealed that the highest saving energy in conventional system belongs to diesel fuel and irrigation inputs, and the least amount of energy saving was seen in human labor input. While for the conservation system, the highest and the least amount of energy saving belongs to nitrogen and human labor, respectively.

Building medical image classifiers with very limited data using segmentation networks.

Deep learning has shown promising results in medical image analysis, however, the lack of very large annotated datasets confines its full potential. Although transfer learning with ImageNet pre-trained classification models can alleviate the problem, constrained image sizes and model complexities can lead to unnecessary increase in computational cost and decrease in performance. As many common morphological features are usually shared by different classification tasks of an organ, it is greatly beneficial if we can extract such features to improve classification with limited samples. Therefore, inspired by the idea of curriculum learning, we propose a strategy for building medical image classifiers using features from segmentation networks. By using a segmentation network pre-trained on similar data as the classification task, the machine can first learn the simpler shape and structural concepts before tackling the actual classification problem which usually involves more complicated concepts. Using our proposed framework on a 3D three-class brain tumor type classification problem, we achieved 82% accuracy on 191 testing samples with 91 training samples. When applying to a 2D nine-class cardiac semantic level classification problem, we achieved 86% accuracy on 263 testing samples with 108 training samples. Comparisons with ImageNet pre-trained classifiers and classifiers trained from scratch are presented.

Successful Management of Right Ventricular Perforation Associated with a Pacemaker Lead During Off-Pump CABG Surgery: A Case Report.

INTRODUCTION: Intraoperative right ventricular perforation due to pacing catheter after its successful and uneventful insertion is a rare complication. Here, we present a case of cardiac arrest due to right ventricular perforation associated with a pacemaker lead during off-pump coronary artery bypass graft surgery. CASE PRESENTATION: The case was a 68-year-old male, who was admitted to our hospital with retrosternal chest pain. He had a history of implantation of a permanent pacemaker due to symptomatic complete atrioventricular block. Based on angiography, the diagnosis was 3- vessel disease involving the left anterior descending, second obtuse marginal, and right coronary arteries. The right ventricle was perforated by the tip of the permanent pacemaker lead during off-pump coronary artery bypass graft surgery. Subsequently, the patient suddenly experienced cardiac arrest and underwent emergency on-pump cardiac surgery. CONCLUSIONS: This case showed that in some situations, emergency surgery as a life saving procedure may be required in cardiac perforation due to permanent pacemaker lead even following cardiac arrest.

Learning in data-limited multimodal scenarios: Scandent decision forests and tree-based features.

Incomplete and inconsistent datasets often pose difficulties in multimodal studies. We introduce the concept of scandent decision trees to tackle these difficulties. Scandent trees are decision trees that optimally mimic the partitioning of the data determined by another decision tree, and crucially, use only a subset of the feature set. We show how scandent trees can be used to enhance the performance of decision forests trained on a small number of multimodal samples when we have access to larger datasets with vastly incomplete feature sets. Additionally, we introduce the concept of tree-based feature transforms in the decision forest paradigm. When combined with scandent trees, the tree-based feature transforms enable us to train a classifier on a rich multimodal dataset, and use it to classify samples with only a subset of features of the training data. Using this methodology, we build a model trained on MRI and PET images of the ADNI dataset, and then test it on cases with only MRI data. We show that this is significantly more effective in staging of cognitive impairments compared to a similar decision forest model trained and tested on MRI only, or one that uses other kinds of feature transform applied to the MRI data.

Prevalence of Coronary Artery Ectasia with Atherosclerosis and Associated Risk Factors in the West of Iran: A Cross-Sectional Study.

BACKGROUND: According to the angiographic findings, 3%-8% of atherosclerotic coronary artery patients suffer from coronary artery ectasia (CAE). We conducted this study to estimate the prevalence of CAE among patients who underwent angiography and compared this group with those patients without CAE and atherosclerosis in terms of common coronary heart disease (CHD) risk factors. METHODS: This cross sectional study was conducted in Hamadan Province, western Iran, from March 2014 to March 2015. Data were collected from angiography evaluation and clinical records in Ekbatan Hospital. The patients with atherosclerosis who had CAE were compared with patients who had neither CAE nor atherosclerosis. The categorical variables were compared using chi-squared test or Fisher's exact test. RESULTS: Of 2767 patients who underwent coronary angiography, 166 (6.0%) had CAE with atherosclerosis, 2357 (85.2%) had atherosclerosis without CAE, and 244 (8.8%) had normal coronary artery. Compared to normal group, CAE patient were more hypertensive and smoker. Besides, in CAE group the proportion of dyslipidemia was higher than normal subject. CONCLUSIONS: The prevalence of CAE in Hamadan Province was in the expected level. Distribution of common CHD risk factors were most prevalent in CAE patient in comparison with normal coronary artery group.

Relation between the serum ferritin level and the risk for acute myocardial infarction.

BACKGROUND: Increased estimated body iron stores have been recently suggested to be associated with increased risk of acute myocardial infarction (AMI); however the question of whether serum ferritin level as an indicator for estimating body iron is an independent risk factor for cardiac events is still questioned. In the present study, we assessed whether serum ferritin was associated with the incidence of AMI. METHODS: The study population consisted of 100 consecutive male patients with first AMI (50 suffered STEMI and 50 with NSTEMI diagnosis) admitted within 12 hours of the onset of chest pain to coronary care units (CCU) at Ekbatan hospital in the city of Hamadan, Iran. A control group (n = 50) was also randomly selected among men without any evidences of AMI from same hospital. Serum ferritin was measured using an ELISA assay by a special kit at the first and fifth days after admission. RESULTS: The first and fifth day serum ferritin concentrations averaged 56.75 and 112.5 µg/dl in STEMI (ST Elevation Myocardial Infarction) group, 36.5 and 87.25 µg/dl in NSTEMI (Non ST Elevation Myocardial Infarction)group and 22.5 and 42.0 µg/dl in control group that was significantly higher in former group. In this regard, the medium level of ferritin in STEMI, NSTEMI, and control groups were 159, 146, and 32.5 µg/dl, respectively that was significantly higher in those who suffered STEMI than in other study subgroups (p < 0.001). Multivariable logistic regression model showed that the elevated level of serum ferritin could predict occurrence of STEMI adjusted for initial ferritin concentration, patients' age and coronary disease risk factors (OR = 5.1, P = 0.017). CONCLUSIONS: Elevated serum ferritin can be a potent factor for predicting AMI especially STEMI.

The frequency of using herbal medicines among patients with hypertension in Kerman, Iran, 2012-2013.

INTRODUCTION: The use of medicinal plants has dramatically increased in recent years. Given the increasing rate of hypertension and medical plants usage by these patients and considering drug interactions due to concomitant use with drugs, the present study aims to evaluate the rate of medicinal plants usage in hypertensive patients. METHODS: This is a cross-sectional study (descriptive-analytical) in which 650 hypertensive patients referring to the subspecialty clinic of Kerman were questioned about medicinal plants usage by a medicinal plants questionnaire. Among these patients, there were 612 who consented to participate. After the variables were described, the data were finally analyzed using Stata 12. RESULTS: The average age of those using these drugs in the past year was 58.8 ± 10 years. Of the total number of participants using medicinal plants, there were 58 males (23.5%) and 122 females (33.4%). There were 129 participants (72.5%) using medicinal plants through self-administration, 17 participants (9.5%) on experienced users' advice, 16 participants (9%) as administered by herbalists, and 11 participants (6%) as administered by doctors. However, the most important resources for using a drug that prevents hypertension were family and friends (74 participants; 41.5%) and doctors (13 participants; 7.3%). According to the results, there was no significant difference between the level of education and medicinal plants usage (P = .95); however, there was a significant difference between gender and medicinal plants usage (P = .009). DISCUSSION: According to the results indicating the relatively high prevalence of medicinal plants usage and their arbitrary use by hypertensive patients without consulting a specialist, it seems necessary to plan for more effective and secure public education and train people to provide herbal drug services for various diseases with hypertension being the most common one.

Augmenting MRI-transrectal ultrasound-guided prostate biopsy with temporal ultrasound data: a clinical feasibility study.

PURPOSE: In recent years, fusion of multi-parametric MRI (mp-MRI) with transrectal ultrasound (TRUS)-guided biopsy has enabled targeted prostate biopsy with improved cancer yield. Target identification is solely based on information from mp-MRI, which is subsequently transferred to the subject coordinates through an image registration approach. mp-MRI has shown to be highly sensitive to detect higher-grade prostate cancer, but suffers from a high rate of false positives for lower-grade cancer, leading to unnecessary biopsies. This paper utilizes a machine-learning framework to further improve the sensitivity of targeted biopsy through analyzing temporal ultrasound data backscattered from the prostate tissue. METHODS: Temporal ultrasound data were acquired during targeted fusion prostate biopsy from suspicious cancer foci identified in mp-MRI. Several spectral features, representing the signature of backscattered signal from the tissue, were extracted from the temporal ultrasound data. A supervised support vector machine classification model was trained to relate the features to the result of histopathology analysis of biopsy cores obtained from cancer foci. The model was used to predict the label of biopsy cores for mp-MRI-identified targets in an independent group of subjects. RESULTS: Training of the classier was performed on data obtained from 35 biopsy cores. A fivefold cross-validation strategy was utilized to examine the consistency of the selected features from temporal ultrasound data, where we achieved the classification accuracy and area under receiver operating characteristic curve of 94 % and 0.98, respectively. Subsequently, an independent group of 25 biopsy cores was used for validation of the model, in which mp-MRI had identified suspicious cancer foci. Using the trained model, we predicted the tissue pathology using temporal ultrasound data: 16 out of 17 benign cores, as well as all three higher-grade cancer cores, were correctly identified. CONCLUSION: The results show that temporal analysis of ultrasound data is potentially an effective approach to complement mp-MRI-TRUS-guided prostate cancer biopsy, specially to reduce the number of unnecessary biopsies and to reliably identify higher-grade cancers.

Ultrasound RF time series for classification of breast lesions.

This work reports the use of ultrasound radio frequency (RF) time series analysis as a method for ultrasound-based classification of malignant breast lesions. The RF time series method is versatile and requires only a few seconds of raw ultrasound data with no need for additional instrumentation. Using the RF time series features, and a machine learning framework, we have generated malignancy maps, from the estimated cancer likelihood, for decision support in biopsy recommendation. These maps depict the likelihood of malignancy for regions of size 1 mm(2) within the suspicious lesions. We report an area under receiver operating characteristics curve of 0.86 (95% confidence interval [CI]: 0.84%-0.90%) using support vector machines and 0.81 (95% CI: 0.78-0.85) using Random Forests classification algorithms, on 22 subjects with leave-one-subject-out cross-validation. Changing the classification method yielded consistent results which indicates the robustness of this tissue typing method. The findings of this report suggest that ultrasound RF time series, along with the developed machine learning framework, can help in differentiating malignant from benign breast lesions, subsequently reducing the number of unnecessary biopsies after mammography screening.