Prediction of melanoma Breslow thickness using deep transfer learning algorithms.

BACKGROUND: The distinction between in situ melanoma (MIS) and invasive melanoma is challenging even for expert dermatologists. The use of pretrained convolutional neural networks (CNNs) as ancillary decision systems needs further research. AIM: To develop, validate and compare three deep transfer learning (DTL) algorithms to predict MIS vs. invasive melanoma and melanoma with a Breslow thickness (BT) of < 0.8 mm vs. ≥ 0.8 mm. METHODS: A dataset of 1315 dermoscopic images of histopathologically confirmed melanomas was created from Virgen del Rocio University Hospital and open repositories of the International Skin Imaging Collaboration archive and Polesie S et al. (Dermatol Pract Concept 2021; 11:e2021079). The images were labelled as MIS or invasive melanoma and < 0.8 mm or ≥ 0.8 mm of BT. We conducted three trainings, and overall means for receiver operating characteristic (ROC) curves, sensitivity, specificity, positive and negative predictive value, and balanced diagnostic accuracy outcomes were evaluated on the test set with ResNetV2, EfficientNetB6 and InceptionV3. The results of 10 dermatologists were compared with the algorithms. Grad-CAM gradient maps were generated, highlighting relevant areas considered by the CNNs within the images. RESULTS: EfficientNetB6 achieved the highest diagnostic accuracy for the comparison between MIS vs. invasive melanoma (61%) and BT < 0.8 mm vs. ≥ 0.8 mm (75%). For the BT comparison, ResNetV2 with an area under the ROC curve of 0.76 and InceptionV3 with an area under the ROC curve of 0.75, outperformed the results obtained by the dermatologist group with an area under the ROC curve of 0.70. CONCLUSION: EfficientNetB6 recorded the best prediction results, outperforming the dermatologists for the comparison of 0.8 mm of BT. DTL could be an ancillary aid to support dermatologists' decisions in the near future.

COVID-19 hospitalized patients and oral changes: a case-control study.

OBJECTIVES: To investigate the association between hospitalization for COVID-19 and oral changes, and to evaluate whether oral changes can indicate a higher risk of disease progression to death. MATERIALS AND METHODS: This case-control study analyzed patients hospitalized (university hospital), including those in intensive care unit and clinical wards. The study group comprised 69 COVID-19 positive patients (PCR-test), while the control group included 43 COVID-19 negative patients. A dentist performed oral evaluations, and salivary samples were collected for calcium, phosphatase, and pH analysis. Sociodemographic data, hospitalization information, and hematological test results were collected from electronic-medical records. The presence of oral changes was assessed using chi-square tests, and the predicted risk of death was analyzed using binary logistic regression. RESULTS: COVID-19 positive patients had a significantly higher prevalence of oral changes compared to COVID-19 negative patients. The presence of any oral changes in COVID-19 positive patients indicated a 13-fold higher risk of mortality. "Bleeding ulcers," "pressure ulcers," and "angular cheilitis" were significantly associated with hospitalization for COVID-19. CONCLUSION: There may be an association between hospitalization for COVID-19 and the development of oral changes, including bleeding ulcers, pressure ulcers. and angular cheilitis. These oral changes may serve as potential indicator for disease progression an increased risk of death. CLINICAL RELEVANCE: COVID-19 hospitalized patients have a higher prevalence of oral changes, which indicate an increased risk of mortality. Oral medicine staff should be included in multidisciplinary teams to detect and treat these oral changes promptly.

Serum s100b protein levels as a neuroinflammatory biomarker of acutely relapsed paranoid schizophrenia patients.

OBJECTIVE: The s100b inflammatory protein is involved in schizophrenia pathophysiology. We aim at studying the evolution of the s100b serum levels in acutely relapsed paranoid schizophrenia patients at three different time points (admission, discharge and 3 months after hospital discharge 3MAHD). METHODS: Twenty-three paranoid schizophrenia inpatients meeting DSM-IV criteria participated in the research. Twenty-three healthy subjects matched by age, gender and season acted as the control group. Psychopathology was measured with the Positive and Negative Syndrome Scale (PANSS). Serum s100b levels were determined at 12:00 and 24:00 h with an enzyme-linked immunoassay kit. RESULTS: Patients had significant higher serum s100b levels at admission and discharge (12:00 h) than the group of healthy subjects. At admission and discharge, s100b serum levels at 24 h had decreased compared to the 24:00 h s100b levels of the healthy subjects. At 3MAHD patients and healthy subjects had similar levels of serum s100b protein. Positive and negative PANSS scores decreased significantly between admission and discharge. Positive and negative PANSS scores decreased between discharge and 3MAHD, but these changes had no statistical significance. CONCLUSIONS: Our study confirms that the acute inflammatory response produced in acutely relapsed patients is reversed after 3 month of hospital discharge. The variations of serum s100b concentrations when the patients suffer from an acute relapse may be a useful predictor of disease evolution.

Peptides, Exopolysaccharides, and Short-Chain Fatty Acids from Fermented Milk and Perspectives on Inflammatory Bowel Diseases.

Crohn's disease and ulcerative colitis are characterized by chronic inflammatory processes and an imbalanced immune response along the gastrointestinal (GI) tract. Pharmacological treatments have been widely used, although their long-term application has adverse side effects. On the other hand, milks fermented with specific lactic acid bacteria (LAB) have been shown to be useful as alternative or complementary aids. Many metabolites such as peptides, exopolysaccharides, and short-chain fatty acids are produced during milk fermentation. These components have been shown to change the pH of the gastrointestinal lumen, aid intestine mucosal recovery, modulate the microbiota, and reduce the inflammatory response (innate and adaptive immune system), both in vitro and in vivo. Therefore, the objective of the present review is to describe how these bioactive compounds from fermented milk by specific LAB can decrease the deleterious symptoms of inflammatory bowel disease.

Recommendations for screening, monitoring, prevention, and prophylaxis of infections in adult and pediatric patients receiving CAR T-cell therapy: a position paper.

Chimeric antigen receptor (CAR) T-cell therapy is one of the most promising emerging treatments for B-cell malignancies. Recently, two CAR T-cell products (axicabtagene ciloleucel and tisagenlecleucel) have been approved for patients with aggressive B-cell lymphoma and acute lymphoblastic leukemia; many other CAR-T constructs are in research for both hematological and non-hematological diseases. Most of the patients receiving CAR-T therapy will develop fever at some point after infusion, mainly due to cytokine release syndrome (CRS). The onset of CRS is often indistinguishable from an infection, which makes management of these patients challenging. In addition to the lymphodepleting chemotherapy and CAR T cells, the treatment of complications with corticosteroids and/or tocilizumab increases the risk of infection in these patients. Data regarding incidence, risk factors and prevention of infections in patients receiving CAR-T cell therapy are scarce. To assist in patient care, a multidisciplinary team from hospitals designated by the Spanish Ministry of Health to perform CAR-T therapy prepared these recommendations. We reviewed the literature on the incidence, risk factors, and management of infections in adult and pediatric patients receiving CAR-T cell treatment. Recommendations cover different areas: monitoring and treatment of hypogammaglobulinemia, prevention, prophylaxis, and management of bacterial, viral, and fungal infections as well as vaccination prior and after CAR-T cell therapy.

Performance Evaluation of Deep Learning-Based Prostate Cancer Screening Methods in Histopathological Images: Measuring the Impact of the Model's Complexity on Its Processing Speed.

Prostate cancer (PCa) is the second most frequently diagnosed cancer among men worldwide, with almost 1.3 million new cases and 360,000 deaths in 2018. As it has been estimated, its mortality will double by 2040, mostly in countries with limited resources. These numbers suggest that recent trends in deep learning-based computer-aided diagnosis could play an important role, serving as screening methods for PCa detection. These algorithms have already been used with histopathological images in many works, in which authors tend to focus on achieving high accuracy results for classifying between malignant and normal cases. These results are commonly obtained by training very deep and complex convolutional neural networks, which require high computing power and resources not only in this process, but also in the inference step. As the number of cases rises in regions with limited resources, reducing prediction time becomes more important. In this work, we measured the performance of current state-of-the-art models for PCa detection with a novel benchmark and compared the results with PROMETEO, a custom architecture that we proposed. The results of the comprehensive comparison show that using dedicated models for specific applications could be of great importance in the future.

Wildlife Monitoring on the Edge: A Performance Evaluation of Embedded Neural Networks on Microcontrollers for Animal Behavior Classification.

Monitoring animals' behavior living in wild or semi-wild environments is a very interesting subject for biologists who work with them. The difficulty and cost of implanting electronic devices in this kind of animals suggest that these devices must be robust and have low power consumption to increase their battery life as much as possible. Designing a custom smart device that can detect multiple animal behaviors and that meets the mentioned restrictions presents a major challenge that is addressed in this work. We propose an edge-computing solution, which embeds an ANN in a microcontroller that collects data from an IMU sensor to detect three different horse gaits. All the computation is performed in the microcontroller to reduce the amount of data transmitted via wireless radio, since sending information is one of the most power-consuming tasks in this type of devices. Multiples ANNs were implemented and deployed in different microcontroller architectures in order to find the best balance between energy consumption and computing performance. The results show that the embedded networks obtain up to 97.96% ± 1.42% accuracy, achieving an energy efficiency of 450 Mops/s/watt.

Evaluation of acrylamide-removing properties of bacterial consortia under simulated gastrointestinal conditions.

BACKGROUND: Previous studies have demonstrated the acrylamide-removing properties of probiotic monocultures; however, potential advantages of consortia over monocultures in reducing the dietary exposure to acrylamide have not been proven. Hence this work aims to assess the acrylamide (AA)-binding properties of bacterial consortia, consisting of either probiotic strains and / or representative bacteria of duodenal microbiota, exposed to simulated gastrointestinal conditions (SGC). The AA binding capacity of ten probiotic strains (PS) and six duodenal strains (NDS) was evaluated under different conditions; then, three different consortia (PS, NDS, and PS + NDS) were assessed under SGC. RESULTS: Among individual PS, Bacillus coagulans GBI-30, Lactobacillus fermentum J23, L. pentosus J37 and J24, and L. casei Shirota, exhibited the highest AA-binding capacity (80-87%), while Bifidobacterium catenulatun ATCC27676, Streptococcus salivarius subsp. thermophilus ATCC19258, and S. gallolyticus ATCC9809 were the best (ca. 68%) NDS monocultures. Probiotic strain consortia showed higher (P < 0.05) AA binding capacity (> 90%) than monoculture bacteria. Conversely, individual NDS cultures displayed higher (P < 0.05) binding capacity than NDS consortia (60%). A significant reduction (P < 0.05) in AA removal capacity was observed when consortia were exposed to SGC, PS consortia being the most effective (> 60% removal). CONCLUSION: These results suggest that consortia of specific PS could play an important role in reducing the intestinal availability of acrylamide. © 2021 Society of Chemical Industry.

Acute paranoid schizophrenia relapsed inpatients present summer/ winter but not day/night changes in serum S100B concentrations.

Healthy subjects present higher summer than winter S100B protein concentrations. There is no available information regarding if schizophrenia patients present the same pattern. The aim of this research is to study if patients with schizophrenia present seasonal changes in serum S100B concentrations.

A 5-year multicentre randomized controlled trial comparing personalized, frozen and fresh blastocyst transfer in IVF.

RESEARCH QUESTION: Does clinical performance of personalized embryo transfer (PET) guided by endometrial receptivity analysis (ERA) differ from frozen embryo transfer (FET) or fresh embryo transfer in infertile patients undergoing IVF? DESIGN: Multicentre, open-label randomized controlled trial; 458 patients aged 37 years or younger undergoing IVF with blastocyst transfer at first appointment were randomized to PET guided by ERA, FET or fresh embryo transfer in 16 reproductive clinics. RESULTS: Clinical outcomes by intention-to-treat analysis were comparable, but cumulative pregnancy rate was significantly higher in the PET (93.6%) compared with FET (79.7%) (P = 0.0005) and fresh embryo transfer groups (80.7%) (P = 0.0013). Analysis per protocol demonstrates that live birth rates at first embryo transfer were 56.2% in PET versus 42.4% in FET (P = 0.09), and 45.7% in fresh embryo transfer groups (P = 0.17). Cumulative live birth rates after 12 months were 71.2% in PET versus 55.4% in FET (P = 0.04), and 48.9% in fresh embryo transfer (P = 0.003). Pregnancy rates at the first embryo transfer in PET, FET and fresh embryo transfer arms were 72.5% versus 54.3% (P = 0.01) and 58.5% (P = 0.05), respectively. Implantation rates at first embryo transfer were 57.3% versus 43.2% (P = 0.03), and 38.6% (P = 0.004), respectively. Obstetrical outcomes, type of delivery and neonatal outcomes were similar in all groups. CONCLUSIONS: Despite 50% of patients dropping out compared with 30% initially planned, per protocol analysis demonstrates statistically significant improvement in pregnancy, implantation and cumulative live birth rates in PET compared with FET and fresh embryo transfer arms, indicating the potential utility of PET guided by the ERA test at the first appointment.

Multicenter prospective study of concordance between embryonic cell-free DNA and trophectoderm biopsies from 1301 human blastocysts.

BACKGROUND: The recent identification of embryonic cell-free DNA in spent blastocyst media has opened a new era of possibilities for noninvasive embryo aneuploidy testing in assisted reproductive technologies. Yet, previous studies assessing a limited number of embryos reported variable concordance between embryonic cell-free DNA and trophectoderm biopsies, thus questioning the validity of this approach. OBJECTIVE: This study aimed to evaluate the concordance and reproducibility of testing embryonic cell-free DNA vs trophectoderm DNA obtained from the same embryo in a large sample of human blastocysts and to assess the contribution of the inner cell mass and trophectoderm to embryonic cell-free DNA released to the culture media. STUDY DESIGN: This is an interim analysis of a prospective, observational study among 8 in vitro fertilization centers in 4 continents to assess consistency between noninvasive embryo aneuploidy testing of embryonic cell-free DNA and conventional trophectoderm biopsy. The analysis included 1301 day-6/7 blastocysts obtained in 406 in vitro fertilization cycles from 371 patients aged 20-44 years undergoing preimplantation genetic testing for aneuploidy. Fresh oocytes underwent intracytoplasmic sperm injection or in vitro fertilization. No previous assisted hatching or vitrification was allowed before media collection. Individual spent blastocyst medium was collected from embryos cultured at least 40 hours from day 4. After media collection, conventional preimplantation genetic testing for aneuploidy, comprising trophectoderm biopsy and blastocyst vitrification, was performed. Embryonic cell-free DNA was analyzed blindly after embryo transfer. Inner cell mass and trophectoderm biopsies were also performed in a subset of 81 aneuploid blastocysts donated for research. RESULTS: Embryonic cell-free DNA analyses were 78.2% (866/1108) concordant with the corresponding trophectoderm biopsies. No significant differences were detected among centers ranging from 72.5% to 86.3%. Concordance rates exceeded 86% when all defined steps in the culture laboratory were controlled to minimize the impact of maternal and operator contamination. Sensitivity per center ranged from 76.5% to 91.3% and specificity from 64.7% to 93.3%. The false-negative rate was 8.3% (92/1108), and false-positive rate was 12.4% (137/1108). The 2 fertilization techniques provided similar sensitivity (80.9% vs 87.9%) and specificity (78.6% vs 69.9%). Multivariate analysis did not reveal any bias from patient clinical background, ovarian stimulation protocols, culture conditions, or embryo quality on testing accuracy of concordance. Moreover, concordances of embryonic cell-free DNA with trophectoderm and inner cell mass suggest that the embryonic cell-free DNA originates from both compartments of the human embryo. CONCLUSION: Noninvasive analysis of embryonic cell-free DNA in spent blastocyst culture media demonstrates high concordance with trophectoderm biopsy results in this large multicenter series. A noninvasive approach for prioritizing embryo euploidy offers important advantages such as avoiding invasive embryo biopsy and decreased cost, potentially increasing accessibility for a wider patient population.

Endometrial scratch vs no intervention in egg donation cycles: the ENDOSCRATCH trial protocol.

BACKGROUND: The effects of endometrial scratching (ES) on embryo implantation have been studied for many years. Several studies have shown better outcomes when performed on patients undergoing intrauterine insemination and in vitro fertilization (IVF) cycles, but many other reports have not been able to find these differences. As far as cycles with donor eggs are concerned, reported evidence is scarce. Our aim in this trial is to determine if ES is useful for those patients undergoing IVF cycles with donor eggs, in order to assure a greater homogeneity in embryo quality and endometrial preparation. METHODS: This single centre randomized controlled trial will include patients undergoing an egg donation cycle, meeting the inclusion criteria and who accept to participate in the study. Once informed consent is signed, patients will be randomly allocated to the study arm (group A) and then receive ES in the luteal phase of the cycle prior to embryo transfer, or the control arm (group B) without any intervention. All cycle data will be collected and analyzed to obtain the clinical pregnancy and the live birth rates in the two groups. DISCUSSION: Several studies have tried to determine the effectiveness of an ES in IVF cycles, but it is still unclear due to the heterogeneity of these reports. The aim of this study is to determine if there are differences in clinical pregnancy rate and live birth rate in egg donor cycles, when comparing an ES performed in the preceding luteal phase versus no intervention, given that embryo quality and endometrial preparation protocols will be comparable. TRIAL REGISTRATION: Ethical approval of version 2.0 of this trial was obtained on the 13th January 2017. It was retrospectively registered on the 5th April 2017 as the ENDOSCRATCH Trial (NCT03108157) in ClinicalTrials.gov.

Life expectancy of follicular lymphoma patients in complete response at 30 months is similar to that of the Spanish general population.

The use of immunochemotherapy has improved the outcome of follicular lymphoma (FL). Recently, complete response at 30 months (CR30) has been suggested as a surrogate for progression-free survival. This study aimed to analyse the life expectancy of FL patients according to their status at 30 months from the start of treatment in comparison with the sex and age-matched Spanish general population (relative survival; RS). The training series comprised 263 patients consecutively diagnosed with FL in a 10-year period who needed therapy and were treated with rituximab-containing regimens. An independent cohort of 693 FL patients from the Grupo Español de Linfomas y Trasplante Autólogo de Médula Ósea (GELTAMO) group was used for validation. In the training cohort, 188 patients were in CR30, with a 10-year overall survival (OS) of 53% and 87% for non-CR30 and CR30 patients, respectively. Ten-year RS was 73% and 100%, showing no decrease in life expectancy for CR30 patients. Multivariate analysis indicated that the FL International Prognostic Index was the most important variable predicting OS in the CR30 group. The impact of CR30 status on RS was validated in the independent GELTAMO series. In conclusion, FL patients treated with immunochemotherapy who were in CR at 30 months showed similar survival to a sex- and age-matched Spanish general population.

Invited review: Bioactive compounds produced during cheese ripening and health effects associated with aged cheese consumption.

Traditionally, cheese is manufactured by converting fluid milk to a semisolid mass through the use of a coagulating agent, such as rennet, acid, heat plus acid, or a combination thereof. Cheese can vary widely in its characteristics, including color, aroma, texture, flavor, and firmness, which can generally be attributed to the production technology, source of the milk, moisture content, and length of aging, in addition to the presence of specific molds, yeast, and bacteria. Among the most important bacteria, lactic acid bacteria (LAB) play a critical role during the cheese-making process. In general, LAB contain cell-envelope proteinases that contribute to the proteolysis of cheese proteins, breaking them down into oligopeptides that can be subsequently taken up by cells via specific peptide transport systems or further degraded into shorter peptides and amino acids through the collaborative action of various intracellular peptidases. Such peptides, amino acids, and their derivatives contribute to the development of texture and flavor in the final cheese. In vitro and in vivo assays have demonstrated that specific sequences of released peptides exhibit biological properties including antioxidant, antimicrobial, anti-inflammatory, immunomodulatory, and analgesic/opioid activity, in addition to angiotensin-converting enzyme inhibition and antiproliferative activity. Some LAB also produce functional lipids (e.g., conjugated linoleic acid) with anti-inflammatory and anticarcinogenic activity, synthesize vitamins and antimicrobial peptides (bacteriocins), or release γ-aminobutyric acid, a nonprotein amino acid that participates in physiological functions, such as neurotransmission and hypotension induction, with diuretic effects. This review provides an overview of the main bioactive components present or released during the ripening process of different types of cheese.

A Lightweight Anonymous Client⁻Server Authentication Scheme for the Internet of Things Scenario: LAuth.

The Internet of Things (IoT) connects different kinds of devices into a network, and enables two-way communication between devices. A large amount of data are collected by these devices and transmitted in this network, it is necessary to ensure secure communications between these devices, to make it impossible for an adversary to undermine this communication. To ensure secure communication, many authentication protocols have been proposed, in this study, a fully anonymous authentication scheme for the Internet of things scenario has been proposed, it enables the remote client to anonymously connect to the server and being serviced by the server. The proposed scheme has been verified by AVISPA and BAN Logic, and the result shows that it is safe. Besides, the simulation shows that the proposed scheme is more efficient in computation cost and communication cost.

Risk of, and survival following, histological transformation in follicular lymphoma in the rituximab era. A retrospective multicentre study by the Spanish GELTAMO group.

The diagnostic criteria for follicular lymphoma (FL) transformation vary among the largest series, which commonly exclude histologically-documented transformation (HT) mandatorily. The aims of this retrospective observational multicentre study by the Spanish Grupo Español de Linfoma y Transplante Autólogo de Médula Ósea, which recruited 1734 patients (800 males/934 females; median age 59 years), diagnosed with FL grades 1-3A, were, (i) the cumulative incidence of HT (CI-HT); (ii) risk factors associated with HT; and (iii) the role of treatment and response on survival following transformation (SFT). With a median follow-up of 6·2 years, 106 patients developed HT. Ten-year CI-HT was 8%. Considering these 106 patients who developed HT, median time to transformation was 2·5 years. High-risk FL International Prognostic Index [Hazard ratio (HR) 2·6, 95% confidence interval (CI): 1·5-4·5] and non-response to first-line therapy (HR 2·9, 95% CI: 1·3-6·8) were associated with HT. Seventy out of the 106 patients died (5-year SFT, 26%). Response to HT first-line therapy (HR 5·3, 95% CI: 2·4-12·0), autologous stem cell transplantation (HR 3·9, 95% CI: 1·5-10·1), and revised International Prognostic Index (HR 2·2, 95% CI: 1·1-4·2) were significantly associated with SFT. Response to treatment and HT were the variables most significantly associated with survival in the rituximab era. Better therapies are needed to improve response. Inclusion of HT in clinical trials with new agents is mandatory.

Validation of the NCCN-IPI for diffuse large B-cell lymphoma (DLBCL): the addition of ß2 -microglobulin yields a more accurate GELTAMO-IPI.

The study included 1848 diffuse large B-cell lymphoma (DLBCL)patients treated with chemotherapy/rituximab. The aims were to validate the National Comprehensive Cancer Network International Prognostic Index (NCCN-IPI) and explore the effect of adding high Beta-2 microglobulin (ß2M), primary extranodal presentation and intense treatment to the NCCN-IPI variables in order to develop an improved index. Comparing survival curves, NCCN-IPI discriminated better than IPI, separating four risk groups with 5-year overall survival rates of 93%, 83%, 67% and 49%, but failing to identify a true high-risk population. For the second aim the series was split into training and validation cohorts: in the former the multivariate model identified age, lactate dehydrogenase, Eastern Cooperative Oncology Group performance status, Stage III-IV, and ß2M as independently significant, whereas the NCCN-IPI-selected extranodal sites, primary extranodal presentation and intense treatments were not. These results were confirmed in the validation cohort. The Grupo Español de Linfomas/Trasplante de Médula ósea (GELTAMO)-IPI developed here, with 7 points, significantly separated four risk groups (0, 1-3, 4 or ≥5 points) with 11%, 58%, 17% and 14% of patients, and 5-year overall survival rates of 93%, 79%, 66% and 39%, respectively. In the comparison GELTAMO IPI discriminated better than the NCCN-IPI. In conclusion, GELTAMO-IPI is more accurate than the NCCN-IPI and has statistical and practical advantages in that the better discrimination identifies an authentic high-risk group and is not influenced by primary extranodal presentation or treatments of different intensity.

Food-derived immunomodulatory peptides.

Food proteins contain specific amino acid sequences within their structures that may positively impact bodily functions and have multiple immunomodulatory effects. The functional properties of these specific sequences, also referred to as bioactive peptides, are revealed only after the degradation of native proteins during digestion processes. Currently, milk proteins have been the most explored source of bioactive peptides, which presents an interesting opportunity for the dairy industry. However, plant- and animal-derived proteins have also been shown to be important sources of bioactive peptides. This review summarizes the in vitro and in vivo evidence of the role of various food proteins as sources of immunomodulatory peptides and discusses the possible pathways involving these properties. © 2016 Society of Chemical Industry.

Resilient Wireless Sensor Networks Using Topology Control: A Review.

Wireless sensor networks (WSNs) may be deployed in failure-prone environments, and WSNs nodes easily fail due to unreliable wireless connections, malicious attacks and resource-constrained features. Nevertheless, if WSNs can tolerate at most losing k - 1 nodes while the rest of nodes remain connected, the network is called k - connected. k is one of the most important indicators for WSNs' self-healing capability. Following a WSN design flow, this paper surveys resilience issues from the topology control and multi-path routing point of view. This paper provides a discussion on transmission and failure models, which have an important impact on research results. Afterwards, this paper reviews theoretical results and representative topology control approaches to guarantee WSNs to be k - connected at three different network deployment stages: pre-deployment, post-deployment and re-deployment. Multi-path routing protocols are discussed, and many NP-complete or NP-hard problems regarding topology control are identified. The challenging open issues are discussed at the end. This paper can serve as a guideline to design resilient WSNs.