J Gerontol A Biol Sci Med Sci. 2024 Feb 19:glae055. doi: 10.1093/gerona/glae055. Online ahead of print.

ABSTRACT

BACKGROUND: Body composition can be accurately quantified from abdominal CT exams and is a predictor for the development of aging-related conditions and for mortality. However, reference ranges for CT-derived body composition measures of obesity, sarcopenia, and bone loss have yet to be defined in the general population.

METHODS: We identified a population-representative sample of 4,900 persons aged 20 to 89 years who underwent an abdominal CT exam from 2010 to 2020. The sample was constructed using propensity score matching to an age and sex stratified sample of persons residing in the 27-county region of Southern Minnesota and Western Wisconsin. The matching included race, ethnicity, education level, region of residence and the presence of 20 chronic conditions. We used a validated deep learning based algorithm to calculate subcutaneous adipose tissue area, visceral adipose tissue area, skeletal muscle area, skeletal muscle density, vertebral bone area, and vertebral bone density from a CT abdominal section.

RESULTS: We report CT-based body composition reference ranges on 4,649 persons representative of our geographic region. Older age was associated with a decrease in skeletal muscle area and density, and an increase in visceral adiposity. All chronic conditions were associated with a statistically significant difference in at least one body composition biomarker. The presence of a chronic condition was generally associated with greater subcutaneous and visceral adiposity, and lower muscle density and vertebrae bone density.

CONCLUSIONS: We report reference ranges for CT-based body composition biomarkers in a population-representative cohort of 4,649 persons by age, sex, body mass index, and chronic conditions.

PMID:38373180 | DOI:10.1093/gerona/glae055

Mol Pharm. 2024 Feb 19. doi: 10.1021/acs.molpharmaceut.3c01124. Online ahead of print.

ABSTRACT

Medicinal chemistry and drug design efforts can be assisted by machine learning (ML) models that relate the molecular structure to compound properties. Such quantitative structure-property relationship models are generally trained on large data sets that include diverse chemical series (global models). In the pharmaceutical industry, these ML global models are available across discovery projects as an "out-of-the-box" solution to assist in drug design, synthesis prioritization, and experiment selection. However, drug discovery projects typically focus on confined parts of the chemical space (e.g., chemical series), where global models might not be applicable. Local ML models are sometimes generated to focus on specific projects or series. Herein, ML-based global models, local models, and hybrid global-local strategies were benchmarked. Analyses were done for more than 300 drug discovery projects at Novartis and ten absorption, distribution, metabolism, and excretion (ADME) assays. In this work, hybrid global-local strategies based on transfer learning approaches were proposed to leverage both historical ADME data (global) and project-specific data (local) to adapt model predictions. Fine-tuning a pretrained global ML model (used for weights' initialization, WI) was the top-performing method. Average improvements of mean absolute errors across all assays were 16% and 27% compared with global and local models, respectively. Interestingly, when the effect of training set size was analyzed, WI fine-tuning was found to be successful even in low-data scenarios (e.g., ∼10 molecules per project). Taken together, this work highlights the potential of domain adaptation in the field of molecular property predictions to refine existing pretrained models on a new compound data distribution.

PMID:38373038 | DOI:10.1021/acs.molpharmaceut.3c01124

Int J Geriatr Psychiatry. 2024 Feb;39(2):e6071. doi: 10.1002/gps.6071.

ABSTRACT

BACKGROUND: Geriatric depression and anxiety have been identified as mood disorders commonly associated with the onset of dementia. Currently, the diagnosis of geriatric depression and anxiety relies on self-reported assessments for primary screening purposes, which is uncomfortable for older adults and can be prone to misreporting. When a more precise diagnosis is needed, additional methods such as in-depth interviews or functional magnetic resonance imaging are used. However, these methods can not only be time-consuming and costly but also require systematic and cost-effective approaches.

OBJECTIVE: The main objective of this study was to investigate the feasibility of training an end-to-end deep learning (DL) model by directly inputting time-series activity tracking and sleep data obtained from consumer-grade wrist-worn activity trackers to identify comorbid depression and anxiety.

METHODS: To enhance accuracy, the input of the DL model consisted of step counts and sleep stages as time series data, along with minimal depression and anxiety assessment scores as non-time-series data. The basic structure of the DL model was designed to process mixed-input data and perform multi-label-based classification for depression and anxiety. Various DL models, including the convolutional neural network (CNN) and long short-term memory (LSTM), were applied to process the time-series data, and model selection was conducted by comparing the performances of the hyperparameters.

RESULTS: This study achieved significant results in the multi-label classification of depression and anxiety, with a Hamming loss score of 0.0946 in the Residual Network (ResNet), by applying a mixed-input DL model based on activity tracking data. The comparison of hyper-parameter performance and the development of various DL models, such as CNN, LSTM, and ResNet contributed to the optimization of time series data processing and achievement of meaningful results.

CONCLUSIONS: This study can be considered as the first to develop a mixed-input DL model based on activity tracking data for the multi-label identification of late-life depression and anxiety. The findings of the study demonstrate the feasibility and potential of using consumer-grade wrist-worn activity trackers in conjunction with DL models to improve the identification of comorbid mental health conditions in older adults. The study also established a multi-label classification framework for identifying the complex symptoms of depression and anxiety.

PMID:38372966 | DOI:10.1002/gps.6071

Med Biol Eng Comput. 2024 Feb 19. doi: 10.1007/s11517-024-03051-w. Online ahead of print.

ABSTRACT

In a treatment or diagnosis related to oral health conditions such as oral cancer and oropharyngeal cancer, an investigation of tongue's movements is a major part. In an automatic measurement of such movement, it must first start with a task of tongue segmentation. This paper proposes a solution of tongue segmentation based on a decoder-encoder CNN-based structure i.e., U-Net. However, it could suffer from a problem of feature loss in deep layers. This paper proposes a Deep Upscale U-Net (DU-UNET). An additional up-sampling of the feature map from a contracting path is concatenated to an upper layer of an expansive path, based on an original U-Net structure. The segmentation model is constructed by training DU-UNET on the two publicly available datasets, and transferred to the self-collected dataset of tongue images with five tongue postures which were recorded at a far distance from a camera under a real-world scenario. The proposed DU-UNET outperforms the other existing methods in our literature reviews, with accuracy of 99.2%, mean IoU of 97.8%, Dice score of 96.8%, and Jaccard score of 96.8%.

PMID:38372910 | DOI:10.1007/s11517-024-03051-w

Abdom Radiol (NY). 2024 Feb 19. doi: 10.1007/s00261-024-04225-8. Online ahead of print.

ABSTRACT

A diagnosis of cirrhosis initiates a shift in the management of chronic liver disease and affects the diagnostic workflow and treatment decision of primary liver cancer. Liver biopsy remains the gold standard for cirrhosis diagnosis, but it is invasive and susceptible to sampling bias and observer variability. Various qualitative and quantitative imaging biomarkers based on ultrasound, CT and MRI have been proposed for noninvasive diagnosis of cirrhosis. Qualitative imaging features are easy to apply but have moderate diagnostic sensitivity. Elastography techniques allow quantitative assessment of liver stiffness and are highly accurate for cirrhosis diagnosis. Ultrasound elastography are widely used in clinical practice, while MR elastography has narrower availability. Although not applicable in clinical practice yet, other quantitative imaging features, including liver surface nodularity, linear and volumetric measurement, extracellular volume fraction, liver enhancement on hepatobiliary phase, and parameters derived from diffusion-weighted imaging, can provide additional information of liver morphology, perfusion, and function, thus may increase diagnosis performance. The introduction of radiomics and deep learning has further improved diagnostic accuracy while reducing subjectivity. Several imaging features may also help to assess liver function and outcomes in patients with cirrhosis. In this review, we summarize the qualitative and quantitative imaging biomarkers for noninvasive cirrhosis diagnosis, and the assessment of liver function and outcomes, and discuss the challenges and future directions in this field.

PMID:38372765 | DOI:10.1007/s00261-024-04225-8

Pathologie (Heidelb). 2024 Feb 19. doi: 10.1007/s00292-024-01299-5. Online ahead of print.

ABSTRACT

OBJECTIVE: Artificial intelligence (AI) holds the potential to make significant advancements in pathology. However, its actual implementation and certification for practical use are currently limited, often due to challenges related to model transferability. In this context, we investigate the factors influencing transferability and present methods aimed at enhancing the utilization of AI algorithms in pathology.

MATERIALS AND METHODS: Various convolutional neural networks (CNNs) and vision transformers (ViTs) were trained using datasets from two institutions, along with the publicly available TCGA-MIBC dataset. These networks conducted predictions in urothelial tissue and intrahepatic cholangiocarcinoma (iCCA). The objective was to illustrate the impact of stain normalization, the influence of various artifacts during both training and testing, as well as the effects of the NoisyEnsemble method.

RESULTS: We were able to demonstrate that stain normalization of slides from different institutions has a significant positive effect on the inter-institutional transferability of CNNs and ViTs (respectively +13% and +10%). In addition, ViTs usually achieve a higher accuracy in the external test (here +1.5%). Similarly, we showcased how artifacts in test data can negatively affect CNN predictions and how incorporating these artifacts during training leads to improvements. Lastly, NoisyEnsembles of CNNs (better than ViTs) were shown to enhance transferability across different tissues and research questions (+7% Bladder, +15% iCCA).

DISCUSSION: It is crucial to be aware of the transferability challenge: achieving good performance during development does not necessarily translate to good performance in real-world applications. The inclusion of existing methods to enhance transferability, such as stain normalization and NoisyEnsemble, and their ongoing refinement, is of importance.

PMID:38372762 | DOI:10.1007/s00292-024-01299-5

J Sci Food Agric. 2024 Feb 19. doi: 10.1002/jsfa.13396. Online ahead of print.

ABSTRACT

BACKGROUND: Quick and accurate detection of nutrient buds is critical for yield prediction and field management in tea plantations. However, the complexity of tea plantation environments and the similarity in color between nutrient buds and older leaves make locating tea nutrient buds challenging.

RESULTS: This research presents a lightweight and efficient detection T-YOLO model for accurately detecting tea nutrient buds in unstructured environments. First, a lightweight module C2fG2 and an efficient feature extraction module DBS are introduced into the backbone and neck of the YOLOv5 baseline model. Second, the head network of the model is pruned to further achieve lightweighting. Finally, the dynamic detection head is integrated to mitigate the feature loss caused by lightweighting. The experimental data show that T-YOLO achieves a mean average precision (mAP) of 84.1%, the parameters are 11.26 M, and the floating-point operations per second (FLOPs) are 17.2 G. Compared to the baseline YOLOv5 model, T-YOLO reduces parameters by 47% and lowers FLOPs by 65%. Additionally, T-YOLO outperforms the existing optimal detection YOLOv8 model by 7.5% in terms of mAP.

CONCLUSION: The T-YOLO model proposed in this study performs well in detecting small tea nutrient buds. It provides a decision-making basis for tea farmers to realize the fine management of smart tea gardens. Additionally, the T-YOLO model outperforms mainstream detection models on the public dataset GWHD, which offers a reference for the detection of other small target crops. This article is protected by copyright. All rights reserved.

PMID:38372581 | DOI:10.1002/jsfa.13396

J Sci Food Agric. 2024 Feb 19. doi: 10.1002/jsfa.13393. Online ahead of print.

ABSTRACT

BACKGROUND: Tea garden pest control is a crucial aspect of ensuring tea quality. In this context, the time series prediction of insect pests in tea gardens holds paramount significance. However, deep learning-based time series prediction techniques are rapidly advancing and research into their use in tea garden pest prediction is limited. The current study investigates the time series of whitefly populations in the Tea Expo Garden, Jurong City, Jiangsu Province employing three deep learning algorithms namely Informer, Long Short-Term Memory Network (LSTM), and LSTM-attention for prediction.

RESULTS: The comparative analysis of the three deep learning algorithms revealed optimal results for the LSTM-attention with an average root mean square error (RMSE) of 2.84 and average mean absolute error (MAE) of 2.52 in 7 days prediction length, respectively. For a prediction length of 3 days, LSTM also achieved the best performance, with an average RMSE of 2.60 and an average MAE of 2.24.

CONCLUSION: These findings suggest that different prediction lengths influence the model performance in tea garden pest time series prediction. Additionally, deep learning could be applied satisfactorily in predicting time series of insect pests in tea gardens based on LSTM-attention. Thus, this study provides a theoretical basis for the research on the time series of pest and disease infestations in tea plants. This article is protected by copyright. All rights reserved.

PMID:38372506 | DOI:10.1002/jsfa.13393

Comput Methods Biomech Biomed Engin. 2024 Feb 19:1-12. doi: 10.1080/10255842.2024.2317438. Online ahead of print.

ABSTRACT

Obstructive sleep apnea (OSA) is associated with various health complications, and snoring is a prominent characteristic of this disorder. Therefore, the exploration of a concise and effective method for detecting snoring has consistently been a crucial aspect of sleep medicine. As the easily accessible data, the identification of snoring through sound analysis offers a more convenient and straightforward method. The objective of this study was to develop a convolutional neural network (CNN) for classifying snoring and non-snoring events based on audio. This study utilized Mel-frequency cepstral coefficients (MFCCs) as a method for extracting features during the preprocessing of raw data. In order to extract multi-scale features from the frequency domain of sound sources, this study proposes the utilization of a multi-branch convolutional neural network (MBCNN) for the purpose of classification. The network utilized asymmetric convolutional kernels to acquire additional information, while the adoption of one-hot encoding labels aimed to mitigate the impact of labels. The experiment tested the network's performance by utilizing a publicly available dataset consisting of 1,000 sound samples. The test results indicate that the MBCNN achieved a snoring detection accuracy of 99.5%. The integration of multi-scale features and the implementation of MBCNN, based on audio data, have demonstrated a substantial improvement in the performance of snoring classification.

PMID:38372231 | DOI:10.1080/10255842.2024.2317438

Sports Biomech. 2024 Feb 19:1-16. doi: 10.1080/14763141.2024.2315243. Online ahead of print.

ABSTRACT

Bodyweight squat is one of the basic sports training exercises. Automatic classification of aberrant squat movements can guide safe and effective bodyweight squat exercise in sports training. This study presents a novel gated long-short term memory with transformer network (GLTN) model for the classification of bodyweight squat movements. Twenty-two healthy young male participants were involved in an experimental study, where they were instructed to perform bodyweight squat in nine different movement patterns, including one acceptable movement defined according to the National Strength and Conditioning Association and eight aberrant movements. Data were acquired from four customised inertial measurement units placed at the thorax, waist, right thigh, and right shank, with a sampling frequency of 200 Hz. The results show that compared to state-of-art deep learning models, our model enhances squat movement classification performance with 96.34% accuracy, 96.31% precision, 96.45% recall, and 96.32% F-score. The proposed model provides a feasible wearable solution to monitoring aberrant squat movements that can facilitate performance and injury risk assessment during sports training. However, this model should not serve as a one-size-fits-all solution, and coaches and practitioners should consider individual's specific needs and training goals when using it.

PMID:38372217 | DOI:10.1080/14763141.2024.2315243

J Dent Res. 2024 Feb 19:220345241226871. doi: 10.1177/00220345241226871. Online ahead of print.

ABSTRACT

Quantitative analysis of irregular anatomical structures is crucial in oral medicine, but clinicians often typically measure only several representative indicators within the structure as references. Deep learning semantic segmentation offers the potential for entire quantitative analysis. However, challenges persist, including segmentation difficulties due to unclear boundaries and acquiring measurement landmarks for clinical needs in entire quantitative analysis. Taking the palatal alveolar bone as an example, we proposed an artificial intelligence measurement tool for the entire quantitative analysis of irregular dental structures. To expand the applicability, we have included lightweight networks with fewer parameters and lower computational demands. Our approach finally used the lightweight model LU-Net, addressing segmentation challenges caused by unclear boundaries through a compensation module. Additional enamel segmentation was conducted to establish a measurement coordinate system. Ultimately, we presented the entire quantitative information within the structure in a manner that meets clinical needs. The tool achieved excellent segmentation results, manifested by high Dice coefficients (0.934 and 0.949), intersection over union (0.888 and 0.907), and area under the curve (0.943 and 0.949) for palatal alveolar bone and enamel in the test set. In subsequent measurements, the tool visualizes the quantitative information within the target structure by scatter plots. When comparing the measurements against representative indicators, the tool's measurement results show no statistically significant difference from the ground truth, with small mean absolute error, root mean squared error, and errors interval. Bland-Altman plots and intraclass correlation coefficients indicate the satisfactory agreement compared with manual measurements. We proposed a novel intelligent approach to address the entire quantitative analysis of irregular image structures in the clinical setting. This contributes to enabling clinicians to swiftly and comprehensively grasp structural features, facilitating the design of more personalized treatment plans for different patients, enhancing clinical efficiency and treatment success rates in turn.

PMID:38372132 | DOI:10.1177/00220345241226871

Anal Methods. 2024 Feb 19. doi: 10.1039/d3ay02062b. Online ahead of print.

ABSTRACT

For spectrometers, baseline drift seriously affects the measurement and quantitative analysis of spectral data. Deep learning has recently emerged as a powerful method for baseline correction. However, the dependence on vast amounts of paired data and the difficulty in obtaining spectral data limit the performance and development of deep learning-based methods. Therefore, we solve these problems from the network architecture and training framework. For the network architecture, a Learned Feature Fusion (LFF) module is designed to improve the performance of U-net, and a three-stage training frame is proposed to train this network. Specifically, the LFF module is designed to adaptively integrate features from different scales, greatly improving the performance of U-net. For the training frame, stage 1 uses airPLS to ameliorate the problem of vast amounts of paired data, stage 2 uses synthetic spectra to further ease reliance on real spectra, and stage 3 uses contrastive learning to reduce the gap between synthesized and real spectra. The experiments show that the proposed method is a powerful tool for baseline correction and possesses potential for application in spectral quantitative analysis.

PMID:38372130 | DOI:10.1039/d3ay02062b

SAR QSAR Environ Res. 2024 Feb 19:1-20. doi: 10.1080/1062936X.2024.2312527. Online ahead of print.

ABSTRACT

The autoignition temperature (AIT) serves as a crucial indicator for assessing the potential hazards associated with a chemical substance. In order to gain deeper insights into model performance and facilitate the establishment of effective methodological practices for AIT predictions, this study conducts a benchmark investigation on Quantitative Structure-Property Relationship (QSPR) modelling for AIT. As novelties of this work, three significant advancements are implemented in the AIT modelling process, including explicit consideration of data quality, utilization of state-of-the-art feature engineering workflows, and the innovative application of graph-based deep learning techniques, which are employed for the first time in AIT prediction. Specifically, three traditional QSPR models (multi-linear regression, support vector regression, and artificial neural networks) are evaluated, alongside the assessment of a deep-learning model employing message passing neural network architecture supplemented by graph-data augmentation techniques.

PMID:38372083 | DOI:10.1080/1062936X.2024.2312527

Heliyon. 2024 Feb 7;10(3):e25714. doi: 10.1016/j.heliyon.2024.e25714. eCollection 2024 Feb 15.

ABSTRACT

BACKGROUND: Agar plate analysis is vital for microbiological testing in industries like food, pharmaceuticals, and biotechnology. Manual inspection is slow, laborious, and error-prone, while existing automated systems struggle with the complexity of real-world agar plates. A shortage of diverse datasets hinders the development and evaluation of robust automated systems.

METHODS: In this paper, two new annotated datasets and a novel methodology for synthetic agar plate generation are presented. The datasets comprise 854 images of cultivated agar plates and 1,588 images of empty agar plates, encompassing various agar plate types and microorganisms. These datasets are an extension of the publicly available BRUKERCOLONY dataset, collectively forming one of the largest publicly available annotated datasets for research. The methodology is based on an efficient image generation pipeline that also simulates cultivation-related phenomena such as haemolysis or chromogenic reactions.

RESULTS: The augmentations significantly improved the Dice coefficient of trained U-Net models, increasing it from 0.671 to 0.721. Furthermore, training the U-Net model with a combination of real and 150% synthetic data demonstrated its efficacy, yielding a remarkable Dice coefficient of 0.729, a substantial improvement from the baseline of 0.518. UNet3+ exhibited the highest performance among the U-Net and Attention U-Net architectures, achieving a Dice coefficient of 0.767.

CONCLUSIONS: Our experiments showed the methodology's applicability to real-world scenarios, even with highly variable agar plates. Our paper contributes to automating agar plate analysis by presenting a new dataset and effective methodology, potentially enhancing fully automated microbiological testing.

PMID:38371986 | PMC:PMC10873726 | DOI:10.1016/j.heliyon.2024.e25714

Heliyon. 2024 Feb 5;10(3):e25574. doi: 10.1016/j.heliyon.2024.e25574. eCollection 2024 Feb 15.

ABSTRACT

Globally, cardiovascular diseases (CVDs) rank among the leading causes of mortality. One out of every three deaths is attributed to cardiovascular disease, according to new World Heart Federation research. Cardiovascular disease can be caused by a number of factors, including stress, alcohol, smoking, a poor diet, inactivity, and other medical disorders like high blood pressure or diabetes. In contrast, for the vast majority of heart disorders, early diagnosis of associated ailments results in permanent recovery. Using newly developed data analysis technology, examining a patient's medical record could aid in the early detection of cardiovascular disease. Recent work has employed machine learning algorithms to predict cardiovascular illness on clinical datasets. However, because of their enormous dimension and class imbalance, clinical datasets present serious issues. An inventive model is offered in this work for addressing these problems. An efficient decision support system, also known as an assistive system, is proposed in this paper for the diagnosis and classification of cardiovascular disorders. It makes use of an optimisation technique and a deep learning classifier. The efficacy of traditional techniques for predicting cardiovascular disease using medical data is anticipated to advance with the combination of the two methodologies. Deep learning systems can reduce mortality rates by predicting cardiovascular illness based on clinical data and the patient's severity level. For an adequate sample size of synthesized samples, the optimisation process chooses the right parameters to yield the best prediction from an enhanced classifier. The 99.58% accuracy was obtained by the proposed method. Also, PSNR, sensitivity, specificity, and other metrics were calculated in this work and compared with systems that are currently in use.

PMID:38371968 | PMC:PMC10873670 | DOI:10.1016/j.heliyon.2024.e25574

Heliyon. 2024 Feb 6;10(3):e25655. doi: 10.1016/j.heliyon.2024.e25655. eCollection 2024 Feb 15.

ABSTRACT

BACKGROUND: Differentiating adrenal adenomas from metastases poses a significant challenge, particularly in patients with a history of extra-adrenal malignancy. This study investigates the performance of three-phase computed tomography (CT) based robust federal learning algorithm and traditional deep learning for distinguishing metastases from benign adrenal lesions.

MATERIAL AND METHODS: This retrospective analysis includes 1187 instances who underwent three-phase CT scans between January 2008 and March 2021, comprising 720 benign lesions and 467 metastases. Utilizing the three-phase CT images, both a Robust Federal Learning Signature (RFLS) and a traditional Deep Learning Signature (DLS) were constructed using the Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression. Their diagnostic capabilities were subsequently validated and compared using metrics such as the Areas Under the Receiver Operating Curve (AUCs), Net Reclassification Improvement (NRI), and Decision Curve Analysis (DCA).

RESULTS: Compared with DLS, the RFLS showed better capability in distinguishing metastases from benign adrenal lesions (average AUC: 0.816 vs.0.798, NRI = 0.126, P < 0.072; 0.889 vs.0.838, NRI = 0.209, P < 0.001; 0.903 vs.0.825, NRI = 0.643, p < 0.001) in the four-testing cohort, respectively. DCA showed that the RFLS added more net benefit than DLS for clinical utility. Moreover, Comparison with state-of-the-art federal learning methods, the results once again confirmed that the RFLS significantly improved the diagnostic performance based on three-phase CT (AUC: AP, 0.727 vs. 0.757 vs. 0.739 vs. 0.796; PCP, 0.781 vs. 0.851 vs. 0.790 vs. 0.882; VP, 0.789 vs. 0.814 vs. 0.779 vs. 0.886).

CONCLUSION: RFLS was superior to DLS for preoperative distinguishing metastases from benign adrenal lesions with multi-phase CT Images.

PMID:38371957 | PMC:PMC10873667 | DOI:10.1016/j.heliyon.2024.e25655

Bioinform Adv. 2024 Jan 20;4(1):vbae010. doi: 10.1093/bioadv/vbae010. eCollection 2024.

ABSTRACT

MOTIVATION: A major challenge in cancer care is that patients with similar demographics, tumor types, and medical histories can respond quite differently to the same drug regimens. This difference is largely explained by genetic and other molecular variabilities among the patients and their cancers. Efforts in the pharmacogenomics field are underway to understand better the relationship between the genome of the patient's healthy and tumor cells and their response to therapy. To advance this goal, research groups and consortia have undertaken large-scale systematic screening of panels of drugs across multiple cancer cell lines that have been molecularly profiled by genomics, proteomics, and similar techniques. These large data drug screening sets have been applied to the problem of drug response prediction (DRP), the challenge of predicting the response of a previously untested drug/cell-line combination. Although deep learning algorithms outperform traditional methods, there are still many challenges in DRP that ultimately result in these models' low generalizability and hampers their clinical application.

RESULTS: In this article, we describe a novel algorithm that addresses the major shortcomings of current DRP methods by combining multiple cell line characterization data, addressing drug response data skewness, and improving chemical compound representation.

AVAILABILITY AND IMPLEMENTATION: MMDRP is implemented as an open-source, Python-based, command-line program and is available at https://github.com/LincolnSteinLab/MMDRP.

PMID:38371918 | PMC:PMC10872075 | DOI:10.1093/bioadv/vbae010

Pol J Radiol. 2024 Jan 22;89:e30-e48. doi: 10.5114/pjr.2024.134817. eCollection 2024.

ABSTRACT

Ovarian cancer poses a major worldwide health issue, marked by high death rates and a deficiency in reliable diagnostic methods. The precise and prompt detection of ovarian cancer holds great importance in advancing patient outcomes and determining suitable treatment plans. Medical imaging techniques are vital in diagnosing ovarian cancer, but achieving accurate diagnoses remains challenging. Deep learning (DL), particularly convolutional neural networks (CNNs), has emerged as a promising solution to improve the accuracy of ovarian cancer detection. This systematic review explores the role of DL in improving the diagnostic accuracy for ovarian cancer. The methodology involved the establishment of research questions, inclusion and exclusion criteria, and a comprehensive search strategy across relevant databases. The selected studies focused on DL techniques applied to ovarian cancer diagnosis using medical imaging modalities, as well as tumour differentiation and radiomics. Data extraction, analysis, and synthesis were performed to summarize the characteristics and findings of the selected studies. The review emphasizes the potential of DL in enhancing the diagnosis of ovarian cancer by accelerating the diagnostic process and offering more precise and efficient solutions. DL models have demonstrated their effectiveness in categorizing ovarian tissues and achieving comparable diagnostic performance to that of experienced radiologists. The integration of DL into ovarian cancer diagnosis holds the promise of improving patient outcomes, refining treatment approaches, and supporting well-informed decision-making. Nevertheless, additional research and validation are necessary to ensure the dependability and applicability of DL models in everyday clinical settings.

PMID:38371888 | PMC:PMC10867948 | DOI:10.5114/pjr.2024.134817

Front Oncol. 2024 Feb 2;14:1346010. doi: 10.3389/fonc.2024.1346010. eCollection 2024.

ABSTRACT

BACKGROUND: Lung cancer (LC) is the second-highest incidence and the first-highest mortality cancer worldwide. Early screening and precise treatment of LC have been the research hotspots in this field. Artificial intelligence (AI) technology has advantages in many aspects of LC and widely used such as LC early diagnosis, LC differential classification, treatment and prognosis prediction.

OBJECTIVE: This study aims to analyze and visualize the research history, current status, current hotspots, and development trends of artificial intelligence in the field of lung cancer using bibliometric methods, and predict future research directions and cutting-edge hotspots.

RESULTS: A total of 2931 articles published between 2003 and 2023 were included, contributed by 15,848 authors from 92 countries/regions. Among them, China (40%) with 1173 papers,USA (24.80%) with 727 papers and the India(10.2%) with 299 papers have made outstanding contributions in this field, accounting for 75% of the total publications. The primary research institutions were Shanghai Jiaotong University(n=66),Chinese Academy of Sciences (n=63) and Harvard Medical School (n=52).Professor Qian Wei(n=20) from Northeastern University in China were ranked first in the top 10 authors while Armato SG(n=458 citations) was the most co-cited authors. Frontiers in Oncology(121 publications; IF 2022,4.7; Q2) was the most published journal. while Radiology (3003 citations; IF 2022, 19.7; Q1) was the most co-cited journal. different countries and institutions should further strengthen cooperation between each other. The most common keywords were lung cancer, classification, cancer, machine learning and deep learning. Meanwhile, The most cited papers was Nicolas Coudray et al.2018.NAT MED(1196 Total Citations).

CONCLUSIONS: Research related to AI in lung cancer has significant application prospects, and the number of scholars dedicated to AI-related research on lung cancer is continually growing. It is foreseeable that non-invasive diagnosis and precise minimally invasive treatment through deep learning and machine learning will remain a central focus in the future. Simultaneously, there is a need to enhance collaboration not only among various countries and institutions but also between high-quality medical and industrial entities.

PMID:38371616 | PMC:PMC10869611 | DOI:10.3389/fonc.2024.1346010

Sci Rep. 2024 Feb 19;14(1):4040. doi: 10.1038/s41598-024-54114-9.

ABSTRACT

Accurately obtaining accurate information about the future traffic flow of all roads in the transportation network is essential for traffic management and control applications. In order to address the challenges of acquiring dynamic global spatial correlations between transportation links and modeling time dependencies in multi-step prediction, we propose a spatial linear transformer and temporal convolution network (SLTTCN). The model is using spatial linear transformers to aggregate the spatial information of the traffic flow, and bidirectional temporal convolution network to capture the temporal dependency of the traffic flow. The spatial linear transformer effectively reduces the complexity of data calculation and storage while capturing spatial dependence, and the time convolutional network with bidirectional and gate fusion mechanisms avoids the problems of gradient vanishing and high computational cost caused by long time intervals during model training. We conducted extensive experiments using two publicly available large-scale traffic data sets and compared SLTTCN with other baselines. Numerical results show that SLTTCN achieves the best predictive performance in various error measurements. We also performed attention visualization analysis on the spatial linear transformer, verifying its effectiveness in capturing dynamic global spatial dependency.

PMID:38369549 | DOI:10.1038/s41598-024-54114-9