From online handwriting to synthetic images for Alzheimer's disease detection using a deep transfer learning approach

Cognitive impairment assessment through handwriting (COGITAT) score: a novel tool that predicts cognitive state from handwriting for forensic and clinical applications

Introduction

Handwriting deteriorates proportionally to the writer's cognitive state. Such knowledge is of special importance in the case of a contested will, where dementia of the testator is claimed, but medical records are often insufficient to decide what the testator's cognitive state really was. By contrast, if the will is handwritten, handwriting analysis allows us to gauge the testator's cognitive state at the precise moment when he/she was writing the will. However, quantitative methods are needed to precisely evaluate whether the writer's cognitive state was normal or not. We aim to provide a test that quantifies handwriting deterioration to gauge a writer's cognitive state.

Methods

We consecutively enrolled patients who came for the evaluation of cognitive impairment at the Outpatient Clinic for Cognitive Impairment of the Department of Neuroscience, Rehabilitation, Ophthalmology, Genetics and Maternal and Child Sciences (DINOGMI) of the University of Genoa, Italy. Additionally, we enrolled their caregivers. We asked them to write a short text by hand, and we administered the Mini Mental State Examination (MMSE). Then, we investigated which handwriting parameters correlated with cognitive state as gauged by the MMSE.

Results

Our study found that a single score, which we called the COGnitive Impairment Through hAndwriTing (COGITAT) score, reliably allows us to predict the writer's cognitive state.

Conclusion

The COGITAT score may be a valuable tool to gage the cognitive state of the author of a manuscript. This score may be especially useful in contested handwritten wills, where clinical examination of the writer is precluded.

An Interpretable Deep Learning Optimized Wearable Daily Detection System for Parkinson's Disease

Walking detection in the daily life of patients with Parkinson's disease (PD) is of great significance for tracking the progress of the disease. This study aims to implement an accurate, objective, and passive detection algorithm optimized based on an interpretable deep learning architecture for the daily walking of patients with PD and to explore the most representative spatiotemporal motor features. Five inertial measurement units attached to the wrist, ankle, and waist are used to collect motion data from 100 subjects during a 10-meter walking test. The raw data of each sensor are subjected to the continuous wavelet transform to train the classification model of the constructed 6-channel convolutional neural network (CNN). The results show that the sensor located at the waist has the best classification performance with an accuracy of 98.01%±0.85% and the area under the receiver operating characteristic curve (AUC) of 0.9981±0.0017 under ten-fold cross-validation. The gradient-weighted class activation mapping shows that the feature points with greater contribution to PD were concentrated in the lower frequency band (0.5~3Hz) compared with healthy controls. The visual maps of the 3D CNN show that only three out of the six time series have a greater contribution, which is used as a basis to further optimize the model input, greatly reducing the raw data processing costs (50%) while ensuring its performance (AUC=0.9929±0.0019). To the best of our knowledge, this is the first study to consider the visual interpretation-based optimization of an intelligent classification model in the intelligent diagnosis of PD.

Linguistic, visuospatial, and kinematic writing characteristics in cognitively impaired patients with beta-amyloid deposition

Introduction

Beta-amyloid (Aβ) deposition, a hallmark of Alzheimer's disease (AD), begins before dementia and is an important factor in mild cognitive impairment (MCI). Aβ deposition is a recognized risk factor for various cognitive impairments and has been reported to affect motor performance as well. This study aimed to identify the linguistic, visuospatial, and kinematic characteristics evident in the writing performance of patients with cognitive impairment (CI) who exhibit Aβ deposition.

Methods

A total of 31 patients diagnosed with amnestic mild cognitive impairment (aMCI) with Aβ deposition, 26 patients with Alzheimer's-type dementia, and 33 healthy control (HC) participants without deposition were administered tasks involving dictation of 60 regular words, irregular words, and non-words consisting of 1-4 syllables. Responses from all participants were collected and analyzed through digitized writing tests and analysis tools.

Results

In terms of linguistic aspects, as cognitive decline progressed, performance in the dictation of irregular words decreased, with errors observed in substituting the target grapheme with other graphemes. The aMCI group frequently exhibited corrective aspects involving letter rewriting during the task. In terms of visuospatial aspects, the AD group displayed more errors in grapheme combination compared to the HC group. Lastly, in the kinematic aspects, both the aMCI group and the AD group exhibited slower writing speeds compared to the HC group.

Discussion

The findings suggest that individuals in the CI group exhibited lower performance in word dictation tasks than those in the HC group, and these results possibly indicate complex cognitive-language-motor deficits resulting from temporal-parietal lobe damage, particularly affecting spelling processing. These results provide valuable clinical insights into understanding linguistic-visuospatial-kinematic aspects that contribute to the early diagnosis of CI with Aβ deposition.

A study of auxiliary screening for Alzheimer’s disease based on handwriting characteristics

Background and objectives

Alzheimer’s disease (AD) has an insidious onset, the early stages are easily overlooked, and there are no reliable, rapid, and inexpensive ancillary detection methods. This study analyzes the differences in handwriting kinematic characteristics between AD patients and normal elderly people to model handwriting characteristics. The aim is to investigate whether handwriting analysis has a promising future in AD auxiliary screening or even auxiliary diagnosis and to provide a basis for developing a handwriting-based diagnostic tool.

Materials and methods

Thirty-four AD patients (15 males, 77.15 ± 1.796 years) and 45 healthy controls (20 males, 74.78 ± 2.193 years) were recruited. Participants performed four writing tasks with digital dot-matrix pens which simultaneously captured their handwriting as they wrote. The writing tasks consisted of two graphics tasks and two textual tasks. The two graphics tasks are connecting fixed dots (task 1) and copying intersecting pentagons (task 2), and the two textual tasks are dictating three words (task 3) and copying a sentence (task 4). The data were analyzed by using Student’s t-test and Mann–Whitney U test to obtain statistically significant handwriting characteristics. Moreover, seven classification algorithms, such as eXtreme Gradient Boosting (XGB) and Logistic Regression (LR) were used to build classification models. Finally, the Receiver Operating Characteristic (ROC) curve, accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and Area Under Curve (AUC) were used to assess whether writing scores and kinematics parameters are diagnostic.

Results

Kinematic analysis showed statistically significant differences between the AD and controlled groups for most parameters (p < 0.05, p < 0.01). The results found that patients with AD showed slower writing speed, tremendous writing pressure, and poorer writing stability. We built statistically significant features into a classification model, among which the model built by XGB was the most effective with a maximum accuracy of 96.55%. The handwriting characteristics also achieved good diagnostic value in the ROC analysis. Task 2 had a better classification effect than task 1. ROC curve analysis showed that the best threshold value was 0.084, accuracy = 96.30%, sensitivity = 100%, specificity = 93.41%, PPV = 92.21%, NPV = 100%, and AUC = 0.991. Task 4 had a better classification effect than task 3. ROC curve analysis showed that the best threshold value was 0.597, accuracy = 96.55%, sensitivity = 94.20%, specificity = 98.37%, PPV = 97.81%, NPV = 95.63%, and AUC = 0.994.

Conclusion

This study’s results prove that handwriting characteristic analysis is promising in auxiliary AD screening or AD diagnosis.

Federated Learning-Based Detection of Invasive Carcinoma of No Special Type with Histopathological Images

Invasive carcinoma of no special type (IC-NST) is known to be one of the most prevalent kinds of breast cancer, hence the growing research interest in studying automated systems that can detect the presence of breast tumors and appropriately classify them into subtypes. Machine learning (ML) and, more specifically, deep learning (DL) techniques have been used to approach this problem. However, such techniques usually require massive amounts of data to obtain competitive results. This requirement makes their application in specific areas such as health problematic as privacy concerns regarding the release of patients’ data publicly result in a limited number of publicly available datasets for the research community. This paper proposes an approach that leverages federated learning (FL) to securely train mathematical models over multiple clients with local IC-NST images partitioned from the breast histopathology image (BHI) dataset to obtain a global model. First, we used residual neural networks for automatic feature extraction. Then, we proposed a second network consisting of Gabor kernels to extract another set of features from the IC-NST dataset. After that, we performed a late fusion of the two sets of features and passed the output through a custom classifier. Experiments were conducted for the federated learning (FL) and centralized learning (CL) scenarios, and the results were compared. Competitive results were obtained, indicating the positive prospects of adopting FL for IC-NST detection. Additionally, fusing the Gabor features with the residual neural network features resulted in the best performance in terms of accuracy, F1 score, and area under the receiver operation curve (AUC-ROC). The models show good generalization by performing well on another domain dataset, the breast cancer histopathological (BreakHis) image dataset. Our method also outperformed other methods from the literature.