The Select and Test Algorithm for Inference in Medical Diagnostic Reasoning: Implementation and Evaluation in Clinical Psychiatry

Transcription

1 The Select and Test Algorithm for Inference in Medical Diagnostic Reasoning: Implementation and Evaluation in Clinical Psychiatry D. A. Irosh P. Fernando School of Electrical Engineering & Computer Science, School of Medicine and Public Health, Faculty of Health and Medicine, University of Newcastle Callaghan, NSW 2308, Australia Frans A. Henskens School of Electrical Engineering & Computer Science; and Priority Research Centre for Health Behaviour University of Newcastle Callaghan, NSW 2308, Australia Abstract Clinical diagnostic reasoning involves an informed search for clinical information driven by diagnostic hypotheses. This is then followed by matching the elicited clinical information with diagnostic criteria for each differential diagnosis resulting in diagnostic conclusions. The existing approaches to clinical reasoning were limited in their capabilities in adequately covering this process, particularly in arriving at diagnostic conclusions. As a solution, this paper presents previously published Select and Test (ST) algorithm that were enhanced with a technique known as orthogonal vector projection method, which is used more efficiently and effectively in arriving diagnostic conclusions. The implementation of the algorithm along with a knowledgebase in psychiatry has been described and the accuracy of the algorithm have been demonstrated by evaluating it using actual patient data. Keywords select and test (ST) alogorithm; orthogonal vector projection method; medical expert systems; automated diagnostic reasoning I. INTRODUCTION Medical diagnostic reasoning involves two stages: 1) search for clinical information, which is driven by diagnostic possibilities; and 2) arriving at diagnostic conclusions based on the collected clinical information. Also the reasoning process utilises an effective representation of knowledge, which can be searched and retrieved efficiently. Various cognitive and external factors (e.g. work environment, time pressure) are known to contribute diagnostic errors that are committed by clinicians [1],[2]. There have been many previous attempts to automate diagnostic reasoning by developing various models and diagnostic algorithms, which include: Parsimonious Covering Theory [3]; Information Processing Approach [4]; Process Model for diagnostic rea-soning [5]; Certainty Factor model [6]; models based on Bayes Theorem [7],[8],[9] and Fuzzy logic [10], [11],[12]; scheme-inductive reasoning [13]; hypotheticodeductive reasoning [14] ; backward and forward reasoning [15]; and pattern recognition [16]. The overall limitation of these approaches was that they do not adequately cover the two stages of diagnostic reasoning, and particularly, an efficient method for achieving the second step is diagnostic reasoning is lacking. This was because the process of arriving at diagnostic conclusions by matching the elicited clinical information with the standard diagnostic criteria for differential diagnoses can be challenging due to complexity resulting from diagnostic criteria. The conventional approaches that have been used to handle this problem include rule-based systems and probabilistic approaches. Nonetheless, rule based systems have difficulties in dealing with missing values and, because of the large number of joint probability distributions that are required in Bayesian approaches, the resulting knowledgebase can become too complicated to manage [17]. Earlier efforts to develop successful medical expert systems such as INTERNIST-1 and CADUCEUS have failed despite more than a decade of development [18]. One of the reasons for such failures can be understood as lack of effective theoretical foundation [19]. In order to fulfil this need the Select and Test (ST) algorithm was introduced [20], which was based on the select and test model (also known as epistemological framework) [21]. However, the initial version of the ST algorithm still lacked an efficient mechanism for dealing with the second stage of diagnostic reasoning. As a remedy to this situation a simple and more efficient technique known as orthogonal vector projection method [22] was implemented and demonstrated in differentiating diagnoses and their severities. Developing successful medical expert systems still needs to overcome other challenges, which include: consideration of the system structure of the organisation where medical expert system is used [23]; and engaging clinicians [24]. Nonetheless, an effective theoretical foundation for diagnostic reasoning can be considered as the initial step towards realisation of successful medical expert systems /16/$31.00 copyright 2016 IEEE ICIS 2016, June 26-29, 2016, Okayama, Japan

2 The following sections describe the inferences in diagnostic reasoning, and how they are used to construct the ST (Select and Test) algorithm [20],[25] followed by its implementation and evaluation using actual patient data. It is important to note that the aim of this endeavour is to demonstrate the effectiveness of ST algorithm as a theoretical foundation for medical diagnostic reasoning, rather than developing a medical expert system in psychiatry. II. INFERENCES IN DIAGNOSTIC REASONING The process of eliciting clinical information from a patient is known as abstraction, which involves a process of transforming what the patient described as symptoms in his or her own terminology to medically defined terms. For example, a patient with depression may describe his mood as a dark cloud which the clinician may translate to depressed mood. With this initial clinical information, a process follows in which diagnostic possibilities (i.e. differential diagnoses) are generated, further clinical information is sought to verify them, and a final diagnostic conclusion is made based on available information. This diagnostic reasoning process is well captured in the Select and Test model [21], which is based on the logical inferences, abduction, deduction, and induction [26]. The following sections describe the implementation of a knowledgebase in psychiatry and evaluation of the diagnostic algorithms using clinical data from psychiatric patients; when reading them it is important to note that the ST model has been adopted in introducing a systematic method of clinical reasoning in psychiatry [27]. The descriptions of the logical inferences, the abstraction, and the ST algorithms, which are given below are kept abstract using set notations as manipulations of different sets. These simply include searching a set for a given element, and adding and removing elements to and from different sets. A preliminary version of the ST algorithm prior to enhancing it with the vector projection method has been previously published as pseudocode using linked lists instead of sets along with its Java coding [20]. Then the enhanced version of ST algorithm with orthogonal vector projection method is to be published elsewhere [25]. A. Abduction Abduction inference involves deriving differential diagnoses based on each elicited symptom. In order to formalise it, let us define as the set of all the symptoms that were found for a patient, and as any given symptom, which is an element of Also, let denote all the diagnoses and denote the posterior probability of diagnosis given the symptom. Diagnoses may vary according to their criticality (e.g. bacterial pneumonia is considered to be more critical than common cold since the former is more life-threatening than the latter). Therefore, each diagnosis is assigned a numerical value according to its criticality. Two threshold values and are used for the probability of diagnosis and criticality respectively. Each diagnosis either with the or is stored in a set named, which will be used in next inference deduction. The algorithm for abduction is described as a subprogram below, and it also uses a set named, which is also described below in deduction. For each For each where If AND, Add to B. Deduction Deduction is the reverse process of abduction, and involves deriving the symptoms that are expected in each differential diagnosis in the set. Any given diagnosis is known to be associated with a set of symptoms. Deduction uses the posterior probability which is the probability of having symptom given diagnosis to select the most likely symptoms that are expected in diagnosis above the threshold value These selected symptoms are stored in the set named Next, each of these selected symptoms need to be elicited, noting the patient may or may not have them. The symptoms that are already elicited in the patient are stored in the set regardless of their presence or absence for the patient. Those symptoms that are present in the patient are stored in the set Once the deduction inference is complete for a diagnosis it is removed from the set and stored in the set, which is used to keep track of all the diagnoses that are considered. The algorithm for deduction is described as a subprogram below. For each For each where If AND, Add to remove from add to

3 C. Induction Induction is the last step of diagnostic reasoning, and involves arriving at the conclusion of likelihood for each diagnosis that were considered and stored based on the clinical information found during abduction and deduction. The likelihood (i.e. how likely each diagnosis is, not to be confused with statistical meaning) of each diagnosis was calculate using the orthogonal vector projection, which is described elsewhere[22]. The expected clinical information for each diagnosis and elicited patient information are represented as two vectors and respectively, and calculating the livelihood of diagnosis involves projecting onto and then deriving the following ratio. For each where, If add to add to. D. Abstraction Abstraction of clinical information from a patient in real life can be a complex process, which not only involves multiple sources of information (e.g. collateral history from family members of patients and other clinicians), intuition, multimodal sensory information (e.g. facial expressions, quality of speech), and emotional status. Therefore, attempts to automate abstraction using technology is challenging and beyond the scope of this research work. As a result, the current state of the ST algorithm relies on the clinician to translate the clinical features described by the patient into the entities in the knowledgebase. Therefore, for the purpose of the algorithm, abstraction is simplified to just checking if the required clinical information is available in a given patient profile as shown below. The patient profile is conceptualised as a set of all the symptoms a patient has, that may not yet be known to the clinician. For each If, add to. Remove from Add to III. THE ST ALGORITHM The ST algorithm was constructed using above described subprograms, and is aimed to achieve an exhaustive search as it iterates abstraction, abduction, and deduction until it completes the search. The ST algorithm is listed below, and has an efficiency of since it has three nested loops. It starts with the input set, which is a subset of consisting of clinical features that patient present with. Begin ; ; ; ; ; For each add to add to ; For each where, add to. While ( AND ) End IV. IMPLEMENTATION The ST algorithm was applied for psychiatric diagnoses, which is a subdomain of clinical medicine, and implemented in Java using Netbeans IDE. Because of the vastness of domain knowledge, knowledge bases can often become very large and require a significant amount of manpower to develop. Also, if the algorithm has to elicit a large amount of clinical information when the knowledgebase is very large, it may not be time-efficient. Based on the conceptualisation that clinical knowledge in psychiatry can be defined as clusters and hierarchies of knowledge items [27], as a compromise, using clinical expertise of the first author who is also a practising psychiatrist and DSM V as the standard diagnostic criteria, a knowledgebase consisting of 70 clinical features 44 psychiatric diagnoses was implemented by clustering related clinical features into a single item. These set of clinical features are listed in Fig.6 of the Appendix. Some psychiatric diagnoses can be considered as a union of two other diagnoses. For example, Panic Disorder With Agoraphobia is a union of Agoraphobia Without a History of Panic Disorder and Panic Disorder Without Agoraphobia. In order to condense the knowledgebase, any diagnoses which was a union of already existing diagnoses were not included.

4 The implementation was simplified by using binary values for both posterior probabilities and representing two knowldgebases, each as a matrix in which rows and columns represent the indices of clinical features and diagnoses respectively. These two knowledgebases are tabulated in Fig. 4 and Fig. 5 of the Appendix. Given a patient profile consisting of clinical features including any presenting symptoms that were found in the patient as the input, the algorithm searches for the other expected symptoms and produces an output including a list of differential diagnoses with their likelihood as shown in Fig. 1. The output also includes all the symptoms that were sought in the patient and the symptoms that were found in patient profile. Fig. 1. Output of the ST-algorithm V. EVALUATION AND RESULTS The Evaluation of the algorithm required a set of clinical data that spans over all given diagnoses reasonably well. Usually, inpatient and outpatient psychiatry setting attracts different diagnoses. For example, often inpatient settings are characterised by more unwell patients with severe conditions such as acute schizophrenia and manic episodes, whereas outpatient settings are characterised by less severe and less acute conditions such as anxiety disorders. Therefore, a total number of 81 patient records, each consisting of set of clinical features and actual diagnosis, were collected from clinical records involving both inpatient and outpatient settings. Some patients had more than one diagnosis resulting in a total number of 122 diagnoses. Ethical approval for use of de-identified patient data were obtained from the Human Research Ethics Committee of Hunter New England Local Area Health District. In order to minimise selection bias, all the patients that were present in an inpatient ward or an outpatient clinic on the day when data were collected were included. Also data were collected from 3 different psychiatric teams each consisting of multidisciplinary team of clinicians including a psychiatrist, psychologists, nurses, and a social worker. The set of clinical features of each patient was given as the input to the algorithm, and the diagnosis given to these patient by their treating psychiatrists based on the documented clinical information was compared with the diagnosis given by the algorithm. In relation to the first stage of clinical reasoning, which involves an informed search driven by diagnostic hypotheses, the ST algorithm was able to elicit all the symptoms in the patient profile for all patients when only the presenting clinical features were given. In relation to the second stage of diagnostic reasoning, a threshold value of 0.5, which is the midpoint in the interval, seemed reasonable intuitively, was used for the diagnostic likelihood in order to determine valid diagnoses. The results including the number of actual cases with each diagnoses, false positives, and false negatives at and are shown in the table in Figure-2. Thresold >=0.5 Thresold >0.5 Index Diagnosis Number of cases False Positives False Negatives False Positives False Negatives 1 Autistic disorder Asperger s disorder Delirium Dementia Alzheimer s type Vascular dementia Alcohol dependence Alcohol intoxication Alcohol abuse THC dependence THC abuse Amphetamine dependence Amphetamine abuse Opiate dependence Opiate abuse Schizophrenia Brief psychotic disorder Delusional disorder THC induced psychotic disorder Amphetamine induced psychotic disorder Major depressive episode Manic episode Major depressive disorder recurrent Dysthymic disorder Bipolar disorder currently depressed Bipolar disorder currently manic Agoraphobia Panic disorder Social anxiety disorder Obsessive compulsive disorder Posttraumatic stress disorder Generalised anxiety disorder Somatization disorder Conversion disorder Hypochondriasis Body dysmorphic disorder Factitious disorder Anorexia nervosa Bulimia nervosa Adjustment disorder Borderline personality disorder Alcohol induced depressive disorder Amphetamine induced depressive disorder Amphetamine induced hypomanic/manic symptoms Attention deficit hyperactivity disorder No diagnosis Fig. 2. Diagnoses and number of cases with each diagnosis It is important to note that the algorithm did not produce any false negatives at. When the threshold value is elevated as is evident when the false positives decreases, but at the same time, false negatives increase. This is an inevitable phenomena observed in any diagnostic tools.

5 However, false positives were mainly observed in diagnoses when their symptoms were a subset of another diagnosis with a larger set of symptoms. For example, the set of symptoms in delusional disorder is a subset of symptoms that are observed in schizophrenia. Similarly, the sets of symptoms in major depressive episode, manic episode, and recurrent major depression, are subsets of the symptom set of bipolar affective disorder. Therefore, when a patient has all or most of the symptoms of bipolar affective disorder, he/she can automatically have all the required symptoms of major depressive episode, manic episode, and recurrent major depression. The same happens in delusional disorder when a patient has all the symptoms of schizophrenia. Therefore, the results were also analysed in relation to diagnostic sensitivity and specificity, after combining these related clinical conditions into major diagnostic categories since it was clinically meaningful to do so. The sensitivity and specificity are two established standard indices of diagnostic abilities of a diagnostic instruments that are used in clinical medicine [28]. This resulted in reduction of false positive cases in psychotic disorders to only 2 and none in mood and anxiety disorder categories. The resulting three major diagnostic groups showed a high level of diagnostic sensitivity and specificity as shown in Fig. 3. Thresold>=0.5 Disorder group Sensitivity(95% CI) Specificity(95% CI) Psychotic disorders (18-22) 100(83.16 to 100)% 98.04(93.10 to 99.76)% Mood disorders( 20-25,39, 40-43) 100(91.19% to 100)% 100(95.60% to 100)% Anxiety disorders(26-31) 100(88.78% to 100)% 100(96.03% to 100)% Fig. 3. Sensitivity and specificity with the 95% confidence interval (CI) for the major diagnostic groups. Index numbers of individual diagnoses in each category are shown within brackets VI. DISCUSSION The False negative results in clinical diagnosis can have serious implications since diagnoses get missed resulting in deprivation of treatment opportunity. Given the small size of the knowledgebase compared to the large number of various clinical features that are observed in patients, it is an important result that there were no any false negatives at. On the other hand, false positives are manageable since the algorithm also produces the true positives simultaneously as a differential diagnosis. Therefore, the true diagnosis is unlikely to be missed since included differential diagnoses can be reviewed and revisited with further clinical input in order to arrive at the final diagnosis. Furthermore, the ST algorithm can be enhanced with the necessary instructions to manage false positive results from the previously described situation. For example, IF, and and are both positive diagnoses, where is the set of symptoms in and is the set of symptoms in, then discard. The ST algorithm implemented in this paper relies on clinicians ability to elicit symptoms from patient, and abstract them to clinically defined items that are used in the knowledgebase. Therefore, the ST algorithm implemented in this paper is to be used under clinician s guidance if it is to be used in real life clinical applications. However, this process can be further automated by breaking down each clinical item in the knowledgebase in to sub items, which can be introduced as an extended layer in the knowledgebase. Furthermore, the knowledgebase can be fine-tuned by replacing binary values of posterior probabilities with more realistic values as part of future enhancement. VII. DISCUSSION This paper has described the ST algorithms for medical diagnostic reasoning including its design, implementation, and evaluation with clinical data. The relatively small knowledgebase in psychiatry, which was implemented for the testing purpose of the algorithm, was demonstrated to be adequate to cover common psychiatric conditions that are presented in inpatient and outpatient settings. The evaluation results demonstrated that the ST algorithm was able to perform both stages of clinical reasoning satisfactorily even with a primitive knowledgebase, which can be enhanced with time through inclusion of more cases. Whilst the process of arriving at diagnostic conclusions by matching the clinical information that was found in patients with standard diagnostic criteria is challenging, it was efficiently dealt with using a relatively simple approach known as the orthogonal vector projection method. However, the orthogonal vector projection method needs to be enhanced when dealing with different diagnoses having overlapping symptoms. With the promising results from evaluation using real patient data (i.e. high diagnostic sensitivity and specificity for major diagnostic categories), it will be possible to introduce the ST algorithms and the knowledgebase for routine clinical work specifically for a preliminary diagnosis or triaging purposes after further enhancements. Importantly, this paper demonstrates the capability of the ST algorithm, which can serve as a better foundation for developing medical expert systems. REFERENCES [1] M. Nendaz and A. Perrier, "Diagnostic errors and flaws in clinical reasoning: mechanisms and prevention in practice," Swiss Med Wkly, vol. 142, p. w13706, [2] I. A. Scott, "Errors in clinical reasoning: causes and remedial strategies," BMJ, vol. 338, :19: [3] [3] J. A. Reggia and Y. Peng, "Modeling diagnostic reasoning: a summary of parsimonious covering theory," Computer Methods and Programs in Biomedicine, vol. 25, pp , [4] P. M. Wortman, "Medical Diagnosis: An Information-Processing Approach," COMPUTERS AND BIOMEDICAL RESEARCH, vol. 5, pp , [5] J. r. Stausberg and M. Person, "A process model of diagnostic reasoning in medicine," International Journal of Medical Informatics, vol. 54, pp. 9-23, [6] E. H. Shortliffe and B. G. Buchanan, "A model of inexact reasoning in medicine," Mathematical Biosciences, vol. 23, pp , 4// 1975.

6 [7] S. Andreassen, F. V. Jensen, and K. G. Olesen, "Medical expert systems based on causal probabilistic networks," International Journal of Bio- Medical Computing, vol. 28, pp. 1-30, 5// [8] T. Chard and E. M. Rubenstein, "A model-based system to determine the relative value of different variables in a diagnostic system using Bayes theorem," International Journal of Bio-Medical Computing, vol. 24, pp , 7// [9] B. S. Todd, R. Stamper, and P. Macpherson, "A probabilistic rule-based expert system," International Journal of Bio-Medical Computing, vol. 33, pp , 9// [10] K. Boegl, K.-P. Adlassnig, Y. Hayashi, T. E. Rothenfluh, and H. Leitich, "Knowledge acquisition in the fuzzy knowledge representation framework of a medical consultation system," Artificial Intelligence in Medicine, vol. 30, pp. 1-26, 1// [11] L. s. Godo, R. L. de Mántaras, J. Puyol-Gruart, and C. Sierra, "Renoir, Pneumon-IA and Terap-IA: three medical applications based on fuzzy logic," Artificial Intelligence in Medicine, vol. 21, pp , 1// [12] T. Vetterlein and A. Ciabattoni, "On the (fuzzy) logical content of CADIAG- 2," Fuzzy Sets and Systems, vol. 161, pp , 7/16/ [13] H. Mandin, A. Jones, W. Woloschuk, and P. Harasym, "Helping students learn to think like experts when solving clinical problems," Academic Medicine, vol. 72, pp , [14] A. S. Elstein, L. S. Shulman, and S. A. Sprafka, Medical Problem-Solving: an Analysis of Clinical Reasoning: Cambridge, MA: Harvard University Press [15] E. Hunt, "Cognitive Science: Definition, Status, and Questions " Annual Review of psychology, vol. 40, pp [16] G. R. Norman, C. L. Coblentz, L. R. Brooks, and C. J. Babcook, "Expertise in visual diagnosis - a review of the literature.," Academic Medicine, vol. 66(suppl), pp. s78-s83, [17] A. Oniésko, P. Lucas, and M. J. Druzdzel, "Comparison of Rule-Based and Bayesian Network Approaches in Medical Diagnostic Systems," in Artificial Intelligence in Medicine: 8th Conference on Artificial Intelligence in Medicine in Europe, AIME 2001 Cascais, Portugal, July 1 4, 2001, Proceedings, S. Quaglini, P. Barahona, and S. Andreassen, Eds., ed Berlin, Heidelberg: Springer Berlin Heidelberg, 2001, pp [18] D.A. Wolfram, " An appraisal of INTERNIST-I " Artificial Intelligence in Medicine, vol. 7, pp , [19] I. Fernando, F. Henskens, and M. Cohen, "A Domain Specific Conceptual Model for a Medical Expert System in Psychiatry, and a Development Framework," in IADIS e-health 2011 International Conference, Rome,Italy, [20] I. Fernando and F. Henskens, "ST Algorithm for Diagnostic Reasoning in Psychiatry," Polibits vol. 48, pp , [21] M. Ramoni, M. Stefanelli, L. Magnani, and G. Barosi, "An epistemological framework for medical knowledge-based systems " IEEE Transactions on Systems, Man and Cybernetics, vol. 22, pp , [22] I. Fernando and F. Henskens, "A Modified Case-Based Reasoning Approach for Triaging Psychiatric Patients Using a Similarity Measure Derived from Orthogonal Vector Projection," in Artificial Life and Computational Intelligence. vol. 8955, S. Chalup, A. Blair, and M. Randall, Eds., ed: Springer International Publishing, 2015, pp [23] R. L. Wears and M. Berg, "Computer Technology and Clinical Work Still Waiting for Godot " JAMA, vol. 293, pp , [24] A. K. Das, "Computers in Psychiatry: A Review of Past Programs and an Analysis of Historical Trends," Psychiatric QUARTERLY, vol. 73, pp , [25] I. Fernando and F. Henskens, "Select and Test (ST) Algorithm for Medical Diagnostic Reasoning," in 17th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD 2016) Shanghai, China, [26] C. S. Peirce, "Illustrations of the logic of science, sixth paper-deduction, induction, hypothesis," The Popular Science Monthly, vol. 1, pp , [27] I. Fernando, M. Cohen, and F. Henskens, "A systematic approach to clinical reasoning in psychiatry," Australasian Psychiatry, vol. 21, pp , June 1, [28] D. G. Altman and J. M. Bland, "Diagnostic tests. 1: Sensitivity and specificity," BMJ : British Medical Journal, vol. 308, pp , 1994.

7 I. APPENDIX Index , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 5 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 6 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 7 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 8 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 9 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 10 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1, 0, 0, 0, 0, 11 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 12 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 13 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 14 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 15 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 16 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 17 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 18 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 19 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 20 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 21 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 22 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 23 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 24 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 25 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 26 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 27 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 28 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 29 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 30 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 31 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 32 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 33 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 34 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 35 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 36 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 37 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 38 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 39 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 40 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 41 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 42 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 43 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 44 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 45 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 46 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 47 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 48 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 49 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 50 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 51 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 52 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 53 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 54 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 55 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 56 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 57 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 58 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 59 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 60 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 61 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 62 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 63 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 64 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 65 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 66 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 67 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 68 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 69 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 70 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, Fig. 4: Knowledgebase-1: representing where and represent row and column numbers.

8 Index , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 1, 2, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0 2 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 1, 0 3 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0 4 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0 5 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 0, 0, 1, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1 6 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0 7 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 8 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 9 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1, 0, 0, 0, 0, , 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, , 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, , 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, , 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, , 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 1, 0, 0, 1, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 1, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, , 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 1, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, , 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 1, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, , 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 1, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, , 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 0, 1, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, , 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 0, 1, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, , 0, 0, 0, 0, 1, 0, 0, 1, 1, 1, 1, 0, 0, 1, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, , 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 1, 1, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, , 0, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, , 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, , 0, 0, 0, 0, 1, 0, 1, 0, 0, 1, 1, 0, 0, 1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, , 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 1, 1, 0, 1, 1, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1, 1, 0, , 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1, 1, 1, 0, , 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, , 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 0, , 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 1, 0, 0, 0, 1, 1, 1, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, , 0, 1, 0, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 1, , 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, , 0, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, , 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 1, 0, 0, 0, 1, 0, 0, 1, 1, 1, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 1, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 1, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 1, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 0, 1, 0, 1, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1 Fig. 5: Knowledgebase-2 representing where and represent row and column numbers.

9 Index Clincal feature 1 Period of depressed mood or psychological distress 2 Period of elevated mood 3 Self-harm/ suicidal behaviour attempt 4 Excessive anxiety or severe agitation 5 Disruptive or disorganised behaviour, overactivity or inattention, 6 Anger and aggression 7 Irrational behaviour or thoughts, perceptual abnormalities, disordered speech 8 Cognitive symptoms (e.g. disorientation, impaired consciousness) 9 Somatic symptoms or bodily concerns 10 Binge eating or concerns about body weight 11 Excessive us use of drug and/or alcohol 12 Impairment in social interactions 13 Impairment in communication and language 14 Restricted, repetitive and stereotyped patterns of behaviour, interests, and activities 15 Memory impairment (gradual) 16 Aphasia, apraxia, agnosia, or impairment in executive functioning 17 Presence of systemic conditions that cause cognitive impairment 18 Evidence of cerebrovascular disease 19 Disturbance in consciousness (i.e. disorientation, memory impairment), which fluctuates 20 Regular use alcohol excessively for a prolonged period with evidence of tolerance or withdrawals 21 Presence of physical, social(e.g. work, relationship) or legal problems due to excessive alcohol use 22 Failed attempts to reduce drinking or previous detox/rehab 23 Evidence of alcohol intoxication 24 Regular use THC excessively for a prolonged period with evidence of tolerance or withdrawals 25 Physical, social(e.g. work, relationship), psychological(e.g. psychotic symptoms) or legal problems due to excessive THC use 26 Regular and excessive use of opiates for a prolonged period with evidence of tolerance or withdrawals 27 Presence of physical, social(e.g. work, relationship) or legal problems due to opiate use 28 Attempts to reduce or give up use of opiates (e.g. previous detox/rehab or methadone treatment) 29 Regular and excessive use of amphetamine related substances for a prolonged period with evidence of tolerance or withdrawals 30 Presence of physical, social(e.g. work, relationship) or legal problems due to amphetamine or related substance use 31 Delusions for nearly 1 month or longer 32 Hallucinations for nearly 1 month or longer 33 Thought disorder or disorganised behaviour including catatonia for one month or longer 34 Negative symptoms( e.g. flat affect, amotivation) for one month or longer 35 delusions, hallucinations or thought disorder lasting for less than one month 36 Onset of delusions, hallucinations or thought disorder temporarily related to use of THC 37 Onset of delusions, hallucinations or thought disorder temporarily related to use of amphetamine 38 Persistent depressed mood or loss of interest/ enjoyment for more than 2 weeks 39 Depressed mood for most days for more than 2 years (not persistent) 40 Loss of appetite or overeating or insomnia or hypersomnia or fatigue 41 Thoughts of guilt or worthlessness or hopelessness or low self-esteem of loss of self-confidence 42 Suicidal thoughts, behaviour or attempts 43 Past history of depression or treatment for depression 44 Elevated or irritable mood or grandiosity 45 Decreased need for sleep or increased activity level or psychomotor agitation 46 Increased speech or racing thoughts 47 Distractibility or impaired concentration 48 Past history of manic episodes or treatment for bipolar affective disorder 49 Recurrent panic attacks with 2 or more symptoms of chest discomfort, nausea, dizziness, feeling dread or loosing control, tingling or hot flushes 50 Anxiety about being in places where panic attacks may occur or avoidance of such places 51 Fear of being humiliated or scrutinised/judged by others and avoidance of social or performance situations 52 Recurrent and persistent obsessions or compulsions 53 Intrusive distressing recollection( e.g. flashbacks, nightmares) of traumatic event or avoidance of its triggers 54 Increased hyperarousal( hypervigilance or exaggerated startle response or impaired concentration or insomnia or irritability) 55 Excessive almost constant worry or apprehensive expectation of worse outcomes about various situations 56 Feeling on edge or muscle tension or restlessness or insomnia 57 Multiple pain, gastrointestinal, sexual and neurological symptoms, which are medically unexplained for several years 58 Medically unexplained unintentionally produced neurological symptoms or deficits (e.g. motor) preceded by a stressor or conflict 59 Preoccupation with fears of having a serious illness for several months 60 Preoccupation with and imagined physical defect or anomaly resulting excessive concern or distress 61 Intentional production of physical or psychological symptoms with the motivation to assume a sick role 62 Intense fear of gaining weight with undue influence of body weight on self-evaluation 63 Weight loss resulting physical symptoms or complications ( amenorrhoea ) 64 Recurrent binge eating with compensatory behaviour ( e.g. induced vomiting, fasting, excessive exercise) and undue influence of body weight on self-evaluation 65 Depressed mood or excessive psychological distress or anxiety or insomnia following a major stressful event and relive if the stressor resolves 66 longstanding affective instability ( rapid fluctuation of mood including anger, low mood and euphoria) 67 Stress-related paranoia or dissociations 68 Recurrent suicidal or self-harm behaviour ( cutting, overdosing) 69 Unstable relationships or fear of abandonment 70 Pervasive hyperavctivity, impulsivity or inattention started before age 7 years Fig. 6: List of clinical features