scholarly journals Deep Learning Applied to Intracranial Hemorrhage Detection

Author(s):  
Luis Cortes-Ferre ◽  
Miguel Angel Gutiérrez-Naranjo ◽  
Juan José Egea-Guerrero ◽  
Marcin Balcerzyk

Intracranial hemorrhage is a serious health problem requiring rapid and often intensive medical care. Identifying the location and type of any hemorrhage present is a critical step in treating the patient. Diagnosis requires an urgent procedure and the detection of the hemorrhage is a hard and time-consuming process for human experts. In this paper, we propose a novel method based on Deep Learning techniques which can be useful as decision support system. Our proposal is two-folded. On the one hand, the proposed technique classifies slices of computed tomography scans for hemorrhage existence or not, achieving 92.7% accuracy and 0.978 ROC-AUC. On the other hand, our method provides visual explanation to the chosen classification by using the so-called Grad-CAM method. TRANSLATE with x English ArabicHebrewPolish BulgarianHindiPortuguese CatalanHmong DawRomanian Chinese SimplifiedHungarianRussian Chinese TraditionalIndonesianSlovak CzechItalianSlovenian DanishJapaneseSpanish DutchKlingonSwedish EnglishKoreanThai EstonianLatvianTurkish FinnishLithuanianUkrainian FrenchMalayUrdu GermanMalteseVietnamese GreekNorwegianWelsh Haitian CreolePersian TRANSLATE with COPY THE URL BELOW Back EMBED THE SNIPPET BELOW IN YOUR SITE Enable collaborative features and customize widget: Bing Webmaster Portal Back TRANSLATE with x English ArabicHebrewPolish BulgarianHindiPortuguese CatalanHmong DawRomanian Chinese SimplifiedHungarianRussian Chinese TraditionalIndonesianSlovak CzechItalianSlovenian DanishJapaneseSpanish DutchKlingonSwedish EnglishKoreanThai EstonianLatvianTurkish FinnishLithuanianUkrainian FrenchMalayUrdu GermanMalteseVietnamese GreekNorwegianWelsh Haitian CreolePersian TRANSLATE with COPY THE URL BELOW Back EMBED THE SNIPPET BELOW IN YOUR SITE Enable collaborative features and customize widget: Bing Webmaster Portal Back

Author(s):  
Sezin Barin ◽  
Murat Saribaş ◽  
Beyza Gülizar Çiltaş ◽  
Gür Emre Güraksin ◽  
Utku Köse

Early diagnosis of intracranial hemorrhage significantly reduces mortality. Hemorrhage is diagnosed by using various imaging methods and the most time-efficient one among them is computed tomography (CT). However, it is clear that accurate CT scans requires time, diligence, and experience. Computer-aided design methods are vital for the treatment because they facilitate early diagnosis of intracranial hemorrhage. At this point, deep learning can provide effective outcomes through an automated diagnosis way. However, as different from the known solutions, diagnosis of five different hemorrhage subtypes is a critical problem to be solved.This study focused on deep learning methods and employed cranial computed tomography scans in order to detect intracranial hemorrhage. The diagnosis approach in the study aimed to detect five subtypes of hemorrhage. In detail, EfficientNet-B3 and ResNet-Inception-V2 architectures were used for diagnosis purposes. Eventually, the study also proposed a two-architecture hybrid method for the diagnosis purpose. The obtained findings by the hybrid method were evaluated in terms of a comparative perspective.Results showed that the newly designed hybrid method was quite effective in terms of increasing classification rates of detecting intracranial hemorrhage according to the subtypes. Briefly, an accuracy of 98.5%, which is higher than those of the EfficientNet-B3 and the Inception-ResNet-V2, were obtained thanks to the developed hybrid method.


IEEE Access ◽  
2021 ◽  
pp. 1-1
Author(s):  
Samira Masoudi ◽  
Sherif Mehralivand ◽  
Stephanie A. Harmon ◽  
Nathan Lay ◽  
Liza Lindenberg ◽  
...  

2021 ◽  
Author(s):  
Francesca Lizzi ◽  
Francesca Brero ◽  
Raffaella Cabini ◽  
Maria Fantacci ◽  
Stefano Piffer ◽  
...  

1980 ◽  
Vol 25 (5) ◽  
pp. 381-385
Author(s):  
Jean-Charles Crombez

The questionnaire on continuing education by the Canadian Psychiatric Association's Council on Education and Professional Liaison, sent in 1978 to all Canadian psychiatrists, raises in the author's mind, in spite of his participation in its establishment, the question of the philosophy behind it. Indeed, seeing signs of a greater problem, he identifies the need for two studies, one dealing with the “object”, the other with the “relationship”. Not elaborating on the first one (description of patients and techniques) which is well known, he describes the second as the knowledge and significance of the encounter (that of two persons inevitably and structurally linked). This “area of relations” paradoxically given too little value in the teaching of psychiatry, is more analogical than logical, more intuitive than deductive, more perceptual than intellectual, and more multifactorial than linear. Yet, this dimension of the encounter (whether individual, familial, group or co-therapy) should take place in conjunction with the objective approach, but the latter occurs alone too often. In order to give to this field of relationship a scientific status of its own, and to reintroduce the techniques instead of using them as guard-rails, proper techniques or methods should be employed or developed if necessary. This includes on the one hand the learning of different levels of awareness and the widening of our perceptual, sensorial, intuitive and analogical capacities. (This would allow for an experience of the fundamental relationship between fields that are apart symptom-wise: dream and awakening, physical and psychic, interior and exterior, fantasy and reality, representations and objects, and so on.) On the other hand this leads us to increase our capacity to listen, to abandon ourselves and to get involved, and to “conceive” a presence within the relationship. Finally, there is this learning of how to observe oneself in a situation, of how to look at what is going on within the encounter (and it is in that very position and this very questioning that the concept of neutrality can be understood, not in the legendary phlegm of impenetrability). This can be done within an “experiential” teaching: for the therapist this means the experience and the study of his own involvement, either with a patient or in groups. Another method is supervision, not as “super”-vision but rather as “inter-discovery” and not as control but rather as “ex-pression.” Working in small groups with colleagues where one can enquire about others’ experiences without any normative goal and with an open attitude is desirable. Another tool would be professional meetings, but not in their current form which is not adapted to the field of the relationship. And so on. The author sees a fundamental necessity for these two fields of the “object” and the “relationship” to be taught conjointly, and neither one nor the other to be excluded from the psychiatrist's training; which is not the case at present. The “field of the object” implies an effort at objectifying, defining variables, causes, using experimental methodology, and a more quantitative analysis. The “field of the relationship” implies positions that are often opposed to this. This contradiction seems necessary and inevitable within every person. One tendency is to make ourselves believe that we avoid this contradiction by pretending to total objectivity: that of scientific psychiatry and clear logic. Finally the author returns to the questionnaire that, precisely in its form, is too uniquely meant for an objective teaching: teaching of diagnoses, illnesses, chart controls, patient controls, teaching through questionnaires, case presentations, putting emphasis on articles or textbooks. This proposed method is adapted for teaching persons considered as entities; and learning techniques considered as reified tools. This is exactly the classical stream of university courses and specialty examinations. This reinforces the illusion. There is also the danger, via the “credit” game, that it will strengthen the already strong tendency to mere objectifying of the subject, of the therapist and of science; that it will privilege a normative vision; and discredit certain essential and humanistic dimensions.


10.2196/26151 ◽  
2021 ◽  
Vol 23 (7) ◽  
pp. e26151
Author(s):  
Stanislav Nikolov ◽  
Sam Blackwell ◽  
Alexei Zverovitch ◽  
Ruheena Mendes ◽  
Michelle Livne ◽  
...  

Background Over half a million individuals are diagnosed with head and neck cancer each year globally. Radiotherapy is an important curative treatment for this disease, but it requires manual time to delineate radiosensitive organs at risk. This planning process can delay treatment while also introducing interoperator variability, resulting in downstream radiation dose differences. Although auto-segmentation algorithms offer a potentially time-saving solution, the challenges in defining, quantifying, and achieving expert performance remain. Objective Adopting a deep learning approach, we aim to demonstrate a 3D U-Net architecture that achieves expert-level performance in delineating 21 distinct head and neck organs at risk commonly segmented in clinical practice. Methods The model was trained on a data set of 663 deidentified computed tomography scans acquired in routine clinical practice and with both segmentations taken from clinical practice and segmentations created by experienced radiographers as part of this research, all in accordance with consensus organ at risk definitions. Results We demonstrated the model’s clinical applicability by assessing its performance on a test set of 21 computed tomography scans from clinical practice, each with 21 organs at risk segmented by 2 independent experts. We also introduced surface Dice similarity coefficient, a new metric for the comparison of organ delineation, to quantify the deviation between organ at risk surface contours rather than volumes, better reflecting the clinical task of correcting errors in automated organ segmentations. The model’s generalizability was then demonstrated on 2 distinct open-source data sets, reflecting different centers and countries to model training. Conclusions Deep learning is an effective and clinically applicable technique for the segmentation of the head and neck anatomy for radiotherapy. With appropriate validation studies and regulatory approvals, this system could improve the efficiency, consistency, and safety of radiotherapy pathways.


Author(s):  
Hari Kishan Kondaveeti ◽  
Gonugunta Priyatham Brahma ◽  
Dandhibhotla Vijaya Sahithi

Deep learning (DL), a part of machine learning (ML), comprises a contemporary technique for processing the images and analyzing the big data with promising outcomes. Deep learning methods are successfully being used in various sectors to gain better results. Agriculture sector is one of the sectors that could be benefitted from the deep learning techniques since the current agriculture techniques cannot keep up with the rapid growth in population. In this chapter, the recent trends in the applications of deep learning techniques in the agricultural sector and the survey of the research efforts that employ deep learning techniques are going to be discussed. Also, the models that are implemented are going to be analyzed and compared with the other existing models.


2013 ◽  
Vol 40 (6Part31) ◽  
pp. 522-522
Author(s):  
Q Diot ◽  
S Bentzen ◽  
D Palma ◽  
L Marks ◽  
S Senan ◽  
...  

2018 ◽  
Vol 7 (3.27) ◽  
pp. 258 ◽  
Author(s):  
Yecheng Yao ◽  
Jungho Yi ◽  
Shengjun Zhai ◽  
Yuwen Lin ◽  
Taekseung Kim ◽  
...  

The decentralization of cryptocurrencies has greatly reduced the level of central control over them, impacting international relations and trade. Further, wide fluctuations in cryptocurrency price indicate an urgent need for an accurate way to forecast this price. This paper proposes a novel method to predict cryptocurrency price by considering various factors such as market cap, volume, circulating supply, and maximum supply based on deep learning techniques such as the recurrent neural network (RNN) and the long short-term memory (LSTM),which are effective learning models for training data, with the LSTM being better at recognizing longer-term associations. The proposed approach is implemented in Python and validated for benchmark datasets. The results verify the applicability of the proposed approach for the accurate prediction of cryptocurrency price.


2014 ◽  
Vol 543-547 ◽  
pp. 1844-1847
Author(s):  
Si Min Zhu ◽  
Hai Yun Deng ◽  
Kai Zheng ◽  
Hua Mei Li ◽  
Xiao Zhou Chen

It is known that the level of the consistency-order of initial value problem is an important standard to determine whether the constructed methods for solving initial value problem of ODEs is suitable or not. There are two methods to solve the consistency-order of initial value problem in general. The one is using the remainder of integral formula as local truncated error, and the other one is using absolute error as local truncated error. In the paper, we propose a novel method based on Gauss-Legendre quadrature formula. It use the method of the remainder of integral formula as local truncated error exists in most of the literatures, and it will be solved once again for the consistency-order of the constructed methods that exist in currently literatures by using absolute error as local truncated error, and then draw a conclusion that is differ from what has been proved correspondingly.


Sign in / Sign up

Export Citation Format

Share Document