Predicting the pair correlation functions of silicate and borosilicate glasses using machine learning.

Glasses offer a broad range of tunable thermophysical properties that are linked to their compositions. However, it is challenging to establish a universal composition-property relationship of glasses due to their enormous compositions and chemical space. Here, we address this problem and develop a metamodel of the composition-atomistic structure relationship of a class of glassy materials via a machine learning (ML) approach. Within this ML framework, an unsupervised deep learning technique, viz., a convolutional neural network (CNN) autoencoder, and a regression algorithm, viz. random forest (RF), are integrated into a fully automated pipeline to predict the spatial distribution of atoms in a glass. The RF regression model predicts the pair correlation function of a glass in a latent space. Subsequently, the decoder of the CNN converts the latent space representation to the actual pair correlation function of the given glass. The atomistic structures of silicate (SiO2) and sodium borosilicate (NBS) based glasses with varying compositions and dopants are collected from molecular dynamics (MD) simulations to establish and validate this ML pipeline. The model is found to predict the atom pair correlation functions for many unknown glasses very accurately. This method is very generic and can accelerate the design, discovery, and fundamental understanding of the composition-atomistic structure relationship of glasses and other materials.

nanoNET: machine learning platform for predicting nanoparticles distribution in a polymer matrix.

Polymer nanocomposites (PNCs) offer a broad range of thermophysical properties that are linked to their compositions. However, it is challenging to establish a universal composition-property relationship in PNCs due to their wide-ranging composition and chemical space. Here, we address this problem and develop a new method to model the composition-microstructure relation of a PNC through an intelligent machine-learning pipeline named nanoNET. The nanoNET is a nanoparticles (NPs) distribution predictor, built upon computer vision and image recognition concepts. It integrates unsupervised deep learning and regression in a fully automated pipeline. We conduct coarse-grained molecular dynamics simulations of PNCs and utilize the data to establish and validate the nanoNET. Within this framework, a random forest regression model predicts the distribution of NPs in a PNC in a latent space. Subsequently, a convolutional neural network-based decoder converts the latent space representation to the actual radial distribution function (RDF) of NPs in the given PNC. The nanoNET predicts NPs distribution in many unknown PNCs very accurately. This method is very generic and can accelerate the design, discovery, and fundamental understanding of composition-microstructure relationships in PNCs and other molecular systems.

A Case Report of Multiple Renal Calculi Treated With Individualized Homoeopathy.

Background: Mineral buildups called kidney stones can be free-floating or affixed to the renal papillae and occur in the renal calyces and pelvis. A major morbidity is associated with the widespread problem of renal stone disease. Open surgical lithotomy and minimally invasive endourological procedures are now the standard for the management of kidney stone symptoms. However, individualized homeopathy (iHOM) has greatly improved treatment of Multiple Renal Calculi. Methods: In the OPD of Dr. D. Y. Patil Homoeopathic Medical College and Research Centre, a 26-year-old female patient with multiple renal calculus was treated homeopathically from August 2021 to January 2022. During the follow-up visits outcome was assessed. To assess whether the changes were due to homoeopathic medicine, an assessment using the modified Naranjo criteria was performed. Results: Over an observation period of 5 months beneficial result from iHOM medicine was seen and so can be used by the physicians in the treatment of Renal Calculi as a complementary health practice. Conclusion: Based on the totality of symptoms, individualized Homoeopathic Medicine (iHOM) Lycopodium Clavatum 30C was given and worked well to dissolve and expel all renal stones. Hence, homeopathy is effective in the fragmentation and ejection of renal calculi and remains one of the most popular treatments for urological problems.

Using handpicked features in conjunction with ResNet-50 for improved detection of COVID-19 from chest X-ray images.

Coronaviruses are a family of viruses that majorly cause respiratory disorders in humans. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a new strain of coronavirus that causes the coronavirus disease 2019 (COVID-19). WHO has identified COVID-19 as a pandemic as it has spread across the globe due to its highly contagious nature. For early diagnosis of COVID-19, the reverse transcription-polymerase chain reaction (RT-PCR) test is commonly done. However, it suffers from a high false-negative rate of up to 67% if the test is done during the first five days of exposure. As an alternative, research on the efficacy of deep learning techniques employed in the identification of COVID-19 disease using chest X-ray images is intensely pursued. As pneumonia and COVID-19 exhibit similar/ overlapping symptoms and affect the human lungs, a distinction between the chest X-ray images of pneumonia patients and COVID-19 patients becomes challenging. In this work, we have modeled the COVID-19 classification problem as a multiclass classification problem involving three classes, namely COVID-19, pneumonia, and normal. We have proposed a novel classification framework which combines a set of handpicked features with those obtained from a deep convolutional neural network. The proposed framework comprises of three modules. In the first module, we exploit the strength of transfer learning using ResNet-50 for training the network on a set of preprocessed images and obtain a vector of 2048 features. In the second module, we construct a pool of frequency and texture based 252 handpicked features that are further reduced to a set of 64 features using PCA. Subsequently, these are passed to a feed forward neural network to obtain a set of 16 features. The third module concatenates the features obtained from first and second modules, and passes them to a dense layer followed by the softmax layer to yield the desired classification model. We have used chest X-ray images of COVID-19 patients from four independent publicly available repositories, in addition to images from the Mendeley and Kaggle Chest X-Ray Datasets for pneumonia and normal cases. To establish the efficacy of the proposed model, 10-fold cross-validation is carried out. The model generated an overall classification accuracy of 0.974 ± 0.02 and a sensitivity of 0.987 ± 0.05, 0.963 ± 0.05, and 0.973 ± 0.04 at 95% confidence interval for COVID-19, normal, and pneumonia classes, respectively. To ensure the effectiveness of the proposed model, it was validated using an independent Chest X-ray cohort and an overall classification accuracy of 0.979 was achieved. Comparison of the proposed framework with state-of-the-art methods reveal that the proposed framework outperforms others in terms of accuracy and sensitivity. Since interpretability of results is crucial in the medical domain, the gradient-based localizations are captured using Gradient-weighted Class Activation Mapping (Grad-CAM). In summary, the results obtained are stable over independent cohorts and interpretable using Grad-CAM localizations that serve as clinical evidence.

Thermodynamic and Kinetic Studies of the Interaction of Nuclear Receptor Coactivator-4 (NCOA4) with Human Ferritin.

Ferritinophagy is a ferritin autophagic degradation process mediated by the selective nuclear receptor coactivator-4 (NCOA4). NCOA4 binds to ferritin and delivers it to nascent autophagosomes, which then merge with the lysosomes for ferritin degradation and iron release. Earlier studies have demonstrated a specific association of NCOA4 with ferritin H-subunits, but not L-subunits. However, neither the thermodynamics of this interaction nor the effect of NCOA4 on iron oxidation, iron mineral core formation, or iron mobilization in ferritin has been explored. Using isothermal titration calorimetry, light absorption spectroscopy, and a soluble fragment (residues 383-522) of human NCOA4 expressed in Escherichia coli, we show that the NCOA4 fragment specifically binds H-rich ferritins with a binding stoichiometry of ∼8 NCOA4 molecules per ferritin shell, and Kd values of ∼0.4 and ∼2 µM for homopolymer H-chain ferritin and heteropolymer H-rich ferritin, respectively. The binding reaction was both enthalpically and entropically favored. Whereas the iron oxidation kinetics were not affected by the presence of NCOA4, iron mobilization from ferritin by two different reducing agents (FMN/NADH and sodium dithionite) showed a strong inhibitory effect that was dependent on the concentration of NCOA4 present in solution. Our results suggest that the binding of NCOA4 to ferritin may interfere in the electron transfer pathway through the ferritin shell and may have important biological implications on cellular iron homeostasis.

Accuracy of machine learning in the diagnosis of odontogenic cysts and tumors: a systematic review and meta-analysis.

BACKGROUND: The recent impact of artificial intelligence in diagnostic services has been enormous. Machine learning tools offer an innovative alternative to diagnose cysts and tumors radiographically that pose certain challenges due to the near similar presentation, anatomical variations, and superimposition. It is crucial that the performance of these models is evaluated for their clinical applicability in diagnosing cysts and tumors. METHODS: A comprehensive literature search was carried out on eminent databases for published studies between January 2015 and December 2022. Studies utilizing machine learning models in the diagnosis of odontogenic cysts or tumors using Orthopantomograms (OPG) or Cone Beam Computed Tomographic images (CBCT) were included. QUADAS-2 tool was used for the assessment of the risk of bias and applicability concerns. Meta-analysis was performed for studies reporting sufficient performance metrics, separately for OPG and CBCT. RESULTS: 16 studies were included for qualitative synthesis including a total of 10,872 odontogenic cysts and tumors. The sensitivity and specificity of machine learning in diagnosing cysts and tumors through OPG were 0.83 (95% CI 0.81-0.85) and 0.82 (95% CI 0.81-0.83) respectively. Studies utilizing CBCT noted a sensitivity of 0.88 (95% CI 0.87-0.88) and specificity of 0.88 (95% CI 0.87-0.89). Highest classification accuracy was 100%, noted for Support Vector Machine classifier. CONCLUSION: The results from the present review favoured machine learning models to be used as a clinical adjunct in the radiographic diagnosis of odontogenic cysts and tumors, provided they undergo robust training with a huge dataset. However, the arduous process, investment, and certain ethical concerns associated with the total dependence on technology must be taken into account. Standardized reporting of outcomes for diagnostic studies utilizing machine learning methods is recommended to ensure homogeneity in assessment criteria, facilitate comparison between different studies, and promote transparency in research findings.

Screening of potential antiplasmodial agents targeting cysteine protease-Falcipain 2: a computational pipeline.

The spread of antimalarial drug resistance is a substantial challenge in achieving global malaria elimination. Consequently, the identification of novel therapeutic candidates is a global health priority. Malaria parasite necessitates hemoglobin degradation for its survival, which is mediated by Falcipain 2 (FP2), a promising antimalarial target. In particular, FP2 is a key enzyme in the erythrocytic stage of the parasite's life cycle. Here, we report the screening of approved drugs listed in DrugBank using a computational pipeline that includes drug-likeness, toxicity assessments, oral toxicity evaluation, oral bioavailability, docking analysis, maximum common substructure (MCS) and molecular dynamics (MD) Simulations analysis to identify capable FP2 inhibitors, which are hence potential antiplasmodial agents. A total of 45 drugs were identified, which have positive drug-likeness, no toxic features and good bioavailability. Among these, six drugs showed good binding affinity towards FP2 compared to E64, an epoxide known to inhibit FP2. Notably, two of them, Cefalotin and Cefoxitin, shared the highest MCS with E64, which suggests that they possess similar biological activity as E64. In an investigation using MD for 100 ns, Cefalotin and Cefoxitin showed adequate protein compactness as well as satisfactory complex stability. Overall, these computational approach findings can be applied for designing and developing specific inhibitors or new antimalarial agents for the treatment of malaria infections.Communicated by Ramaswamy H. Sarma.

Functional and Radiological Outcomes of Unstable Proximal Femur Fractures Fixed With Anatomical Proximal Locking Compression Plate.

Introduction Peritrochanteric fractures are the most frequent fractures of the proximal femur that accounts for nearly half of all proximal femur fractures. They are a major cause of disability in the elderly. The aim is to study the functional and radiological outcome of unstable proximal femur fractures fixed with proximal femur locking compression plate (PF-LCP) and its complications. Unstable proximal femur fracture patients operated with proximal femur locking compression plate were followed up functionally by Harris Hip Score and radiologically by neck-shaft angle measure. Materials and methods A retrospective analysis of 30 patients with unstable peritrochanteric fractures treated with PF-LCP in the first-level trauma center was conducted between 2015 and 2019. Stable peritrochanteric, pediatric and open fractures, and polytrauma were excluded. As a mid-term follow-up, functional and radiological outcomes were assessed at six weeks, three months, six months, and 12 months. Data was analyzed using a chi-square test, and results were compared with available western literature.  Results Thirty patients with unstable peritrochanteric fractures operated between 2015 and 2019, complying with our inclusion criteria, were analyzed. All patients were operated by the same surgeon and were available for a mid-term follow-up (12 months). Mean radiological union time was 12.5+/-2 weeks, with 24 patients achieving union between 10-15 weeks, three patients had union little more than 15 weeks. Two patients had non-union and required re-surgery. Functional results were assessed in the 30 patients available for follow-up using Harris Hip Score. Excellent results were seen in 17, good in seven, fair in three, and poor in three patients. Conclusions The choice of implant used to manage unstable peritrochanteric fractures has always been a debatable subject in our orthopedic fraternity. In our study, we used the anatomic, fixed-angle plates in peritrochanteric fractures and obtained significant functional and radiological outcomes over a midterm follow-up. We recommend PF-LCP as a good, stable alternative in the treatment of peritrochanteric femoral fractures. We consider that fracture pattern and extent in the proximal femur have a definite influence in determining the implant of choice. It provides good-to-excellent bone healing with reduced complications and better biomechanical stability.

Evaluation of Results of Ponseti Technique in Idiopathic Clubfoot using Clinical Evaluation and Radiological Assessment.

Background: Congenital clubfoot (congenital talipes equinovarus) occurs in approximately one in 1000 live births and is one of the most common congenital birth defects. The Ponseti method is at present a well-established method of treatment for idiopathic clubfoot deformities. Aim: The aim of the present study was to evaluate the results of serial casting in clubfoot deformity with Ponseti method on the basis of Pirani's scoring and radiological findings before and after completion of treatment. Materials and Methods: A total of 30 patients were enrolled in the study and were treated with Ponseti's casting after grading the severity of deformity clinically by Pirani's scoring and radiological assessment by calculating the talo-first metatarsal angle in anteroposterior (AP) view and talocalcaneal angle in AP and lateral views. The same clinical and radiological assessment was done at the end of treatment before putting a patient on foot abduction orthosis (FAO). Results: The average number of casts applied before full correction was 5.56 (range: 5-8). The average duration of treatment was about 6.65 weeks before the patient was put on FAO. Pirani score significantly improved from an average of 5.50 (range: 4-6) on presentation to 0.24 (range: 0-2) after correction of deformity. Conclusion: The Ponseti method is an excellent method for the correction of all four deformities associated with congenital idiopathic clubfoot, and we found that the addition of radiographic to clinical evaluation helps in the better assessment of correction. It provides statistically significant results both clinically as measured by Pirani severity score and radiologically assessed by talocalcaneal and talo-first metatarsal angle.

Fly Cell Atlas: A single-nucleus transcriptomic atlas of the adult fruit fly.

For more than 100 years, the fruit fly Drosophila melanogaster has been one of the most studied model organisms. Here, we present a single-cell atlas of the adult fly, Tabula Drosophilae, that includes 580,000 nuclei from 15 individually dissected sexed tissues as well as the entire head and body, annotated to >250 distinct cell types. We provide an in-depth analysis of cell type-related gene signatures and transcription factor markers, as well as sexual dimorphism, across the whole animal. Analysis of common cell types between tissues, such as blood and muscle cells, reveals rare cell types and tissue-specific subtypes. This atlas provides a valuable resource for the Drosophila community and serves as a reference to study genetic perturbations and disease models at single-cell resolution.

To Analyze the Role of Intravenous Tranexamic Acid in Hip Fracture surgeries in Orthopedic Trauma.

INTRODUCTION: Hip fractures in orthopedic trauma cases are increasing. Majority of such patients undergoing surgery require blood transfusion of one or more units. Intravenous (I. V.) Tranexamic acid (TXA) may decrease loss of blood, decrease need of blood transfusion, and improve postoperative hemoglobin (Hb) along with lesser adverse effects. Risk of thromboembolic phenomena remains a concern. A study was done to analyze the role of I. V. TXA in hip fracture surgeries in trauma cases. MATERIALS AND METHODS: Sixty patients were included in the study; in two groups (37 males and 23 females), Group A in which two doses of I. V. TXA 15 mg/kg were given and Group B in which two doses of I. V. placebo were given. RESULTS: Total number of randomized hip arthroplasty cases was 22 (11 in Group A and 11 in Group B) whereas randomized osteosynthesis cases were 38 (19 in Group A and 19 in Group B). Mean preoperative Hb value in Group A was 10.8 gm% and in Group B was 10.7 gm% (P > 0.005. Mean postoperative Hb value in Group A was Hb 9.8 gm% and in Group B 9.5 gm% (difference of 3.061%). Mean duration of surgery in Group A was 64.2 min and in Group B was 66.3 min. Mean total blood loss (intraoperative and postoperative) in Group A was 384.6 ml and in Group B was 448.7 ml (14.29% less in Group A). A total of 14 patients in Group A (17 red blood cells [RBCs] units) and 17 patients (21 RBC units) in Group B required RBC transfusion. No major vascular event, severe bacterial infections, symptomatic deep vein thrombosis, pulmonary embolism, limb ischemia, acute coronary syndrome, or immediate postoperative mortality was noted in either group. CONCLUSION: I. V. TXA has the potential to decrease risk of blood transfusion, decrease total blood loss, and to maintain a higher postoperative Hb value with no significant adverse reactions. As the number of cases of hip fractures continues to increase along with increase in age, so the use of TXA in such cases may improve clinical outcomes, lessen number of inpatient days and hence decrease overall cost.

Assessment of health infrastructure in tackling COVID-19: A case study of European and American scenario.

The COVID-19 coronavirus pandemic is an unparalleled threat intoday's quickly developing climate, and we face it as a global community. Like climate change, it is challenging our resilience from environmental health, social security, and government, to knowledge exchange and economic policy in all sectors of the economy and all fields of growth. So much as climate change, everybody's coming together would require the initiative. Throughout Europe and America, several organizations have mobilized to ensure that the neediest are not left behind, encouraging emergencies and disruptions avoidance and preparedness. The coronavirus outbreak has highlighted the growing community's strengths and vulnerabilities that it has influenced, and has provided us with the ability to benefit from each other's accomplishments and shortcomings. The comparison graph has also been shown in this paper displaying European and American scenarios. The globe might feel smaller amid disaster states and global travel bans, but it is a period when teamwork and looking outward were never more relevant.

The Evaluation of the Radiological and Functional Outcome of Distraction Osteogenesis in Patients with Infected Gap Nonunions of Tibia Treated by Bone Transport.

AIM: The aims of this investigation were: 1) to study the Functional outcome of performing distraction osteogenesis in cases of infected non-union of tibia treated with Ilizarov and Limb Reconstruction System, and 2) to study the Radiological outcome of performing distraction osteogenesis in cases of infected non-union of tibia treated with Ilizarov and Limb Reconstruction System. METHODS: The study was done with 27 patients of infected gap nonunions of the tibia at Sir JJ Hospitals, Mumbai from 2013-2016. After implant removal, if required radical resection of necrotic tissue and fractures were stabilised with Ilizarov or mono-lateral fixator depending on non-union site. Corticotomy was either done proximally or distally. Patients were followed up at monthly intervals for a minimum of 6 months. RESULTS: The ASAMI-Bone healing score was Excellent or Good in 86% patients, and Functional score was Excellent or Good in 89% of patients. The commonest problems were of pin tract infection, wire loosening and angulation of the transported segment. CONCLUSION: Elderly age, persistent infection, sensory loss in the foot, the stiffness of the knee, and above all the patient's reluctance to go any further given the protracted treatment besides, systemic disorders such as diabetes are all pointers for considering amputation as an alternative.

Delirium in medical intensive care units: Incidence, subtypes, risk factors, and outcome.

BACKGROUND: Delirium is a frequent yet underdiagnosed neuropsychiatric condition encountered in intensive care units (ICUs). Being both a preventable and potentially reversible process associated with significant morbidity and mortality, understanding risk factors that predispose and precipitate delirium in any given patient are critical in ICUs. AIMS AND OBJECTIVES: The aim of this study is to evaluate the incidence, motor subtypes, risk factors, and clinical outcome of delirium in the medical ICU. MATERIALS AND METHODS: We used a prospective study design on a cohort of consecutive medical ICU admissions of a tertiary care teaching hospital. The Confusion Assessment Method-ICU and Richmond Agitation Sedation Scale were used to diagnose and motor subtype delirium, respectively, along with a checklist to assess risk factors. RESULTS: Of the 280 ICU admissions, 88 (31.4%) developed delirium. Hypoactive delirium was the most common motor subtype (55.7%). The detection rate of delirium was 12.5% (lowest for hypoactive delirium at 2.04%). Age, gender, and years of education did not significantly predict delirium (all P > 0.05). Tobacco use, chronic liver disease, and past episodes of delirium significantly predisposed, while mechanical ventilation, hypoxia, fever, raised levels of bilirubin and creatinine, and benzodiazepine administration significantly precipitated ICU delirium. Delirium was significantly associated with longer ICU stay (t = 4.23, P = 0.000) and 1-month postdischarge mortality (χ 2 = 6.867, P = 0.009). CONCLUSION: Detection of delirium is challenging, especially in ICU patients on mechanical ventilation and hypoactive delirium. Screening and monitoring for predisposing and precipitating risk factors can greatly improve the odds of detection and intervention as ICU delirium is associated with significant morbidity and mortality.

Ferritin exhibits Michaelis-Menten behavior with oxygen but not with iron during iron oxidation and core mineralization.

The excessively high and inconsistent literature values for Km,Fe and Km,O2 prompted us to examine the iron oxidation kinetics in ferritin, the major iron storage protein in mammals, and to determine whether a traditional Michaelis-Menten enzymatic behavior is obeyed. The kinetics of Fe(ii) oxidation and mineralization catalyzed by three different types of ferritins (recombinant human homopolymer 24H, HuHF, human heteropolymer ∼21H:3L, HL, and horse spleen heteropolymer ∼3.3H:20.7L, HosF) were therefore studied under physiologically relevant O2 concentrations, but also in the presence of excess Fe(ii) and O2 concentrations. The observed iron oxidation kinetics exhibited two distinct phases (phase I and phase II), neither of which obeyed Michaelis-Menten kinetics. While phase I was very rapid and corresponded to the oxidation of approximately 2 Fe(ii) ions per H-subunit, phase II was much slower and varied linearly with the concentration of iron(ii) cations in solution, independent of the size of the iron core. Under low oxygen concentration close to physiological, the iron uptake kinetics revealed a Michaelis-Menten behavior with Km,O2 values in the low µM range (i.e. ∼1-2 µM range). Our experimental Km,O2 values are significantly lower than typical cellular oxygen concentration, indicating that iron oxidation and mineralization in ferritin should not be affected by the oxygenation level of cells, and should proceed even under hypoxic events. A kinetic model is proposed in which the inhibition of the protein's activity is caused by bound iron(iii) cations at the ferroxidase center, with the rate limiting step corresponding to an exchange or a displacement reaction between incoming Fe(ii) cations and bound Fe(iii) cations.

Functional Outcome of Elbow Kinematics in Radial Head Excision Versus Radial Head Replacement: A Comparative Study.

BACKGROUND: Radial head fractures are quite common with incidence 1.5-4% of all adult fractures. The treatment for these fractures depends upon age, type of injury and whether the physics is closed or not. AIM: Comparison between radial head excision versus radial head replacement based on mayo elbow scoring in comminuted radial head fractures. MATERIAL AND METHODS: We did a prospective comparative study comprising 32 patients between age 22-60 years with Mason type II/III radial head fractures at Sir J.J Group of Hospitals, Mumbai. The patients were randomised using the admission day of the week placing 17 patients in the arthroplasty group and 15 patients in the excision group. The patients were followed up for 18-24 months (average 20 months) postoperatively. Results were analysed by the Mayo's elbow performance score at 6 months and 18 months and were statistically evaluated by unpaired t-test. RESULTS: At 6 months, radial head arthroplasty gave excellent results in 2 patients, good in 5 patients and fair in 8 patients. In excision, there were 5 patients with excellent results at 6 months, 7 with good results and 2 with fair results. At 18 months, of the 17 patients who had undergone head arthroplasty, 2 had excellent results, and the same number had poor results. 7 (46.7%) of the 15 cases who had undergone radial head excision had excellent results. Good results were obtained in 7 cases of each. There was 6 cases (35.3%) of radial head arthroplasty which fell into the fair group. As per Mayo's score at 6 months follow up, mean and standard deviation (SD) of the scores in arthroplasty was 68.82 and 18.66 respectively & for excision, it was 85.66 and 10.66. At 18 months follow up, it was 75 and 14.89 for arthroplasty & 90.66 and 7.98 for excision. The difference between the results was statistically significant (p < 0.01). CONCLUSION: Our study shows that long and short-term results of radial head excision are better as compared to arthroplasty in comminuted radial head fractures based on mayo elbow scoring, particularly for dominant upper limbs.

DeepFix: A Fully Convolutional Neural Network for Predicting Human Eye Fixations.

Understanding and predicting the human visual attention mechanism is an active area of research in the fields of neuroscience and computer vision. In this paper, we propose DeepFix, a fully convolutional neural network, which models the bottom-up mechanism of visual attention via saliency prediction. Unlike classical works, which characterize the saliency map using various hand-crafted features, our model automatically learns features in a hierarchical fashion and predicts the saliency map in an end-to-end manner. DeepFix is designed to capture semantics at multiple scales while taking global context into account, by using network layers with very large receptive fields. Generally, fully convolutional nets are spatially invariant-this prevents them from modeling location-dependent patterns (e.g., centre-bias). Our network handles this by incorporating a novel location-biased convolutional layer. We evaluate our model on multiple challenging saliency data sets and show that it achieves the state-of-the-art results.